IL-11 Induces Encephalitogenic Th17 Cells in Multiple Sclerosis and Experimental Autoimmune Encephalomyelitis

Key Points

• IL-11⁺CD4⁺ cells accumulate in the CSF and active brain RRMS lesions.

• IL-11 increased the numbers of CNS-infiltrating IL-17⁺CD4⁺ cells in RREAE.

• Passive transfer of IL-11–induced encephalitogenic CD4⁺ T cells induced severe RREAE.

Abstract

IL-11⁺CD4⁺ cells accumulate in the cerebrospinal fluid of patients with early relapsing-remitting multiple sclerosis (MS) and in active brain MS lesions. Mouse studies have confirmed a causal role of IL-11 in the exacerbation of relapsing-remitting experimental autoimmune encephalomyelitis (RREAE). Administration of IL-11 at the time of clinical onset of RREAE induced an acute exacerbation and increased clinical scores, which persisted during the entire course of the disease. IL-11 increased the numbers of spinal cord inflammatory foci, as well as the numbers of peripheral and CNS-infiltrating IL-17⁺CD4⁺ cells and IL-17A serum levels. Ag recall assays revealed that IL-11 induces IL-17A⁺, GM-CSF⁺, and IL-21⁺CD4⁺ myelin Ag-reactive cells. Passive transfer of these encephalitogenic CD4⁺ T cells induced severe RREAE with IL-17A⁺CCR6⁺ CD4⁺ and B cell accumulation within the CNS. Furthermore, passive transfer of nonmanipulated CNS-derived mononuclear cells from mice with RREAE after a single dose of IL-11 induced severe RREAE with increased accumulation of IL-17A⁺ and CCR6⁺ CD4⁺ cells within the CNS. These results suggest that IL-11 might serve as a biomarker of early autoimmune response and a selective therapeutic target for patients with early relapsing-remitting MS.

Introduction

Immunomodulatory therapies are most effective when administered early in the course of relapsing-remitting multiple sclerosis (RRMS). Therefore, we have been searching for biomarkers of the early autoimmune response to accurately identify patients with clinically isolated syndrome (CIS) suggestive of multiple sclerosis (MS), who are amenable to early disease-modifying therapies (1). Our previous study in CIS patients has identified IL-11 as the most significantly increased cytokine in the cerebrospinal fluid (CSF) and serum in comparison with healthy control (HC) subjects. Moreover, IL-11 serum levels were significantly higher in relapses than in the remissions of untreated RRMS patients, suggesting the involvement of this cytokine in the pathogenesis of RRMS. In vitro studies have revealed that IL-11 induces Th17 cell differentiation and expansion in CIS patients (2). Our human studies have identified that CD4⁺ cells represent a predominant source of IL-11 within the peripheral circulation. In comparison with healthy donors, IL-11⁺CD4⁺ cells from CIS patients were significantly increased in the peripheral circulation and exhibited the highest CCR6 expression (86%) among CD4⁺ T cell subsets, which implied their potential for early migration to the CNS (3).

IL-11 is a member of the IL-6 cytokine family, whose prototypical cytokine promotes Th17 differentiation in both mice and humans (4). However, IL-6 alone does not induce Th17 differentiation, in contrast to IL-11, which induces Th17 cell differentiation and expansion that were selectively blocked by αIL-11 mAb and not by αIL-6 mAb (2). These previously reported human in vitro studies prompted current in vivo studies of the causative role of IL-11 in the development of the Th17 autoimmune responses in relapsing-remitting experimental autoimmune encephalomyelitis (RREAE), an animal model of RRMS.

In the presence of the ligand-binding subunits IL-6Rα and IL-11Rα, IL-6 and IL-11 bind to the signal transduction unit gp130 at overlapping epitopes, leading to the formation of ternary complexes with similar downstream signaling (5). STAT3, a transcription factor involved in Th17 differentiation (6), is activated in response to IL-6/IL-6R and IL-11/IL-11R signaling (5). Our laboratory and others have reported that IL-11Rα is expressed by multiple PBMC subsets, with predominant expression in T cells (2, 7).

Zhang et al. (8) reported that IL-11 expression in chronic brain MS lesions is primarily localized to activated astrocytes at the lesion border, whereas IL-11Rα is expressed on oligodendrocytes. However, the inflammatory cells within lesions were not studied. In vitro studies conducted by the same group have demonstrated that IL-11Rα signaling increases oligodendrocyte survival and proliferation via STAT3 phosphorylation (9). Although we acknowledge the reported findings on the role of IL-11 in chronic MS that suggest increased in vitro oligodendrocyte differentiation and survival, we propose that the proinflammatory effect of IL-11 may prevail in the context of active MS lesions and the expression of IL-11R on T cells, monocytes (1), and B cells (10), which exhibit an inflammatory response to IL-11.

In the current study, we found that IL-11⁺CD4⁺ cells are significantly enriched in the CSF of RRMS patients in comparison with their matched blood samples. Immunohistochemistry studies of active brain MS lesion biopsy samples revealed an enrichment of IL-11⁺CD4⁺ cells in comparison with the peripheral circulation, suggesting a role for this cytokine in the development of inflammatory CNS lesions. Animal studies have confirmed the causal role of IL-11 in the induction of acute exacerbations and increased clinical scores throughout the course of RREAE. A single dose of IL-11 significantly increased the number of CNS-infiltrating IL-17A⁺CD4⁺ cells and IL-17A serum levels in comparison with control RREAE mice. Ag recall experiments with spleen and lymph node (LN) cells from proteolipid protein (PLP)_139–151-immunized mice revealed that IL-11 induces encephalitogenic CD4⁺ cells, characterized by IL-17A, GM-CSF, and IL-21 secretion, which, upon passive transfer, induce severe RREAE with IL-17A⁺CCR6⁺ CD4⁺ cell and B cell accumulation within the recipients’ CNS. Furthermore, passive transfer of CNS-derived, nonmanipulated mononuclear cells from mice with active RREAE following a single IL-11 administration induced more-severe disease, with increased accumulation of IL-17A⁺ and CCR6⁺ CD4⁺ cells within the recipient CNS, in comparison with the passive transfer of nonmanipulated CNS-derived cells from RREAE mice.

Materials and Methods

Study subjects

Six RRMS patients were enrolled in the study upon signing an Institutional Review Board–approved consent form. The inclusion criteria for patients consisted of an RRMS diagnosis (11) and age of 27–55 (average 43.3 ± 10.6); all patients were female, with five being white, and one being African American. The exclusion criterion was prior treatment with immunomodulatory or immunosuppressive therapy. Brain biopsy samples were obtained for diagnostic purposes from an additional five untreated RRMS patients, age 20–66 (average 39.6 ± 17.3), under a protocol approved by the UNC Institutional Review Board. Supplemental Table I presents brain magnetic resonance imaging (MRI) and pathology findings from patients that underwent diagnostic brain lesion biopsy.

Active experimental autoimmune encephalomyelitis

All animal procedures were approved by the Institutional Animal Care and Use Committee of the University of North Carolina at Chapel Hill. Eight- to twelve-week-old female SJL/J mice purchased from Charles River Laboratories were sacrificed as HCs or immunized with 80 μg of PLP_139–151 peptide per mouse in CFA containing Mycobacterium tuberculosis (4 mg/ml). Pertussis toxin (200 ng) was injected i.p. on day 0 and day 2, following immunization. Starting from day 12 postimmunization, 12 mice per group were i.p. injected daily with either recombinant mouse IL-11 (0.5 μg/mouse) (Peprotech) or control vehicle (PBS) for 10 d. 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. The mice were monitored for 70 d for the clinical scores, followed by histological studies.

Where indicated, on day 12 and on day 56 postimmunization, six mice per group were i.p. injected with recombinant mouse IL-11 (0.5 μg/mouse) or control vehicle (PBS). After 16 h, the mice were sacrificed. Blood was collected via cardiac puncture, and the mice were perfused with 50 ml of PBS containing heparin (10 U/ml) (Sigma-Aldrich). LNs, including LN1 (superficial cervical, deep cervical, axillary, and brachial LNs) and LN2 (inguinal LNs); spleen; spinal cord; and brain tissue were harvested. Sera were collected for ELISA, and cells were separated from blood and peripheral lymphoid tissues. Spinal cord and brain tissue 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 cell strainer and washed with PBS, followed by 38% Percoll gradient for separation of the CNS mononuclear cells.

Flow cytometry (FACS)

Cells were separated from the blood and CSF samples of RRMS patients or from the blood and tissues of mice with RREAE that received a single injection of IL-11 or control vehicle as well as from the recipients of the LN CD4⁺ cells or CNS-derived mononuclear cells. The cells were stimulated with PMA (50 ng/ml) and ionomycin (500 ng/ml) (Sigma-Aldrich) for 2 h and with BFA (1:1000 dilution) (eBioscience) for an additional 3 h for the intracellular staining, as previously reported (2, 12). The cells were harvested, fixed, permeabilized, and stained with fluorescently conjugated Abs against human IL-11 (R&D Systems), IL-17A, IFN-γ, (eBioscience), and CD4 (BD Bioscience) or murine IL-17A, IFN-γ, IL-4, TNF-α, IL-1β, IL-10, CD11b, CD8a, CD19 (eBioscience), Ly6C, CCR6, ICAM-1, VLA-4 and CD4 (BD Bioscience), GM-CSF (Miltenyi Biotec), IL-21, IL-21R, and IL-6R (R&D Systems). Isotype controls were used to determine the background. The percentage of cells expressing each molecule was determined in gated cells using a BD FACSCalibur Flow Cytometer with CellQuest software (BD Biosciences).

Immunohistochemistry

Brain MS lesion tissue was obtained from five patients using a stereotactic needle biopsy. Sequential dual immunofluorescent staining of brain sections was performed in a Bond fully automated slide staining system (Leica Biosystems) using Bond Research Detection (DS9455) and Bond Polymer Refine Detection (DS9800) kits. Slides were deparaffinized in Bond dewax solution (AR9222) and hydrated in Bond wash solution (AR9590). Mouse mAb against human CD4 (clone 4B12, NCL-L-CD4-368) was from Leica Biosystems (Norwell, MA). Rabbit polyclonal Abs against human IL-11 (sc-7924) and IL-17A (HPA052258) were from Santa Cruz Biotechnology (Dallas, TX) and Sigma-Aldrich (St. Louis, MO), respectively. IL-11 epitope retrieval was done for 30 min at 100°C in Bond epitope retrieval solution 1 (pH 6) (AR9661) and in solution 2 for IL-17A (at pH 9) (AR9640). After pretreatment, slides were incubated for 1 h with anti–IL-11 (1:50) or anti–IL-17A (1:200) mAb followed with Leica goat anti-rabbit polymer DS (RE72290-K) and the tyramide Cy5 (no. SAT705A001EA; PerkinElmer, Boston, MA). The unmasking of CD4 epitope was done for 10 min in Bond epitope retrieval solution 2. Anti-CD4 mAb (1:200) was applied for 30 min, followed with the secondary Ab from a Bond Polymer Refine kit (DS9800) and tyramide Cy3 (SAT704A001EA; Perkin Elmer). Nuclei were stained with Hoechst 33258 (no. H3569; Life Technologies, Grand Island, NY). The stained slides were mounted with ProLong Gold Antifade Reagent (no. P36934; Life Technologies). Digital imaging and analysis were performed using high-resolution acquisition (×20 objective) of the stained slides in the DAPI, Cy3, and Cy5 channels using the Aperio ScanScope FL (Leica Biosystems). Cell nuclei were visualized in DAPI channel (blue), CD4 were visualized in Cy3 channel (green), and IL-11 and IL-17A were visualized in Cy5 (red). For all slides, demyelinated regions on images were annotated by a neuropathologist based on H&E and Luxol fast blue staining. Automated digital analysis of the images was run in both myelinated and demyelinated regions. Tissue Studio software (Tissue Studio version 2.4 with Tissue Studio Library version 4.0; Definiens, Carlsbad, CA) was used to enumerate cells that coexpress the biomarkers of interest in the annotated regions.

For the mouse study, on day 70 postimmunization, six mice per group that received IL-11 or control vehicle were sacrificed, and on day 70 post–passive transfer, six mice per group that received IL-11–polarized CD4⁺ or control Ag-specific CD4⁺ cells were sacrificed, and the brain and spinal cords were removed and fixed in 4% paraformaldehyde. Paraffin sections were stained with Luxol fast blue and H&E. Inflammatory infiltrates were quantified as described in the previous study by a blinded neuropathologist (13).

Western blotting

In an in vitro signaling experiment, naive mice spleen cells were incubated in the absence or presence of IL-11 (R&D Systems) dose titration and αIL-11 mAb at 10 μg/ml (R&D Systems). Total and phosphorylated STAT3 (Cell Signaling), RORγτ, and β actin expression was determined after 30 min.

Cell lysates from the spleen cells derived from four mice with RREAE and four mice with RREAE that received a single dose of IL-11 were denatured in SDS and resolved by 10–15% SDS-PAGE depending on the m.w. of the protein of interest. The proteins were electroblotted onto PVDF membranes (Bio-Rad). The membranes were probed with primary Abs against mouse NF-κB p65 (Cell Signaling), RORγτ (R&D Systems), and β-actin (Santa Cruz Biotechnology), followed by HRP-conjugated IgG secondary Ab (Santa Cruz Biotechnology) incubation. Protein phosphatase 2A inhibitor (Santa Cruz Biotechnology) was added to the sample. The protein expression was detected by Imagequant LAS 4000 software (GE Healthcare Life Sciences). Quantification of band densities was performed using Image Quant TL software (GE Healthcare Life Sciences).

RT-PCR

Cells were separated from the blood and tissues of RREAE mice that received a single dose of IL-11 or control vehicle. The total RNA was extracted, and cDNA was synthesized using a High-Capacity cDNA Archive Kit (Applied Biosystems). The primers for IL-17A, IFN-γ, IL-4, IL-21, IL-22, IL-11, RORc, T-bet, GATA3, AHR, and 18S mRNA were purchased from Applied Biosystems, and gene expression was measured by RT-PCR using TaqMan Gene Expression Assays (Applied Biosystems) in triplicate. The results are expressed as the average relative gene expression and normalized against the 18S mRNA expression.

ELISA

Serum samples were collected from mice with RREAE that received a single dose of IL-11 or control vehicle. The production of IL-11 (R&D Systems) and IL-17A (eBioscience) was measured, as previously reported (2). The sample incubation was extended to 24 h at 4°C, and the detection Ab incubation was prolonged to 2 h at room temperature. The results are expressed for each mouse as the cytokine concentration in picogram per milliliter.

Ag recall assay

Eight- to ten-week-old female SJL/J mice with active experimental autoimmune encephalomyelitis (EAE) were sacrificed at day 11 postimmunization. Spleen and LN cells were harvested and cultured in the absence or presence of PLP_139–151 (20 μg/ml), IL-11 (100 ng/ml), and IL-23 (20 ng/ml) for 7 d. The percentage of cells expressing the indicated intracellular cytokines and surface markers was determined in gated CD4⁺, CD8⁺, and CD19⁺ cells by flow cytometry.

Passive transfer RREAE

LN cells were separated from donor RREAE mice at day 11 postimmunization and cultured in three conditions: 1) PLP_139–151 (20 μg/ml), 2) PLP_139–151 + IL-11 (100 ng/ml), or 3) PLP_139–151 + IL-23 (20 ng/ml) for 7 d. After 7 d culture, CD4⁺ T cells were separated using a CD4⁺ magnetic bead isolation kit (Miltenyi Biotech). A total of 20 × 10⁶ CD4⁺ T cells (purity >98%) per recipient were resuspended in 500 μl 1× PBS for passive transfer via i.p. injection. Clinical scores were assigned daily for 70 d in recipient mice.

Statistical analysis

The results of the clinical scores, FACS, RT-PCR, ELISA, and histology studies are presented as means and, where indicated, SD. They were analyzed using a two-tailed Student paired or unpaired t test. The comparison of multiple groups was performed using a repeated measures ANOVA (GraphPad Software).

Results

IL-11⁺CD4⁺ and Th17 cells are enriched in the CSF and brain lesions of RRMS patients

Our studies revealed that serum IL-11 levels correlate with the concomitant brain MRI T2 and T1 lesion volumes (X. Zhang, unpublished results), supporting the role of this cytokine in CNS lesion formation (14).

We have reported that the percentage of IL-11⁺CD4⁺ cells in the peripheral blood is significantly increased in CIS patients, in comparison with the age-, sex-, and race-matched HCs (2). Those cells highly express CCR6, implying their potential for the migration to the CNS via a similar mechanism as Th17 cells (3). Indeed, the flow cytometry studies of cells derived from the CSF and concomitant blood samples from the six recently diagnosed, untreated RRMS patients (average age 43.3 ± 10.6 y, all female, five white, one African American) revealed that the mean percentage of IL-11⁺CD4⁺, IL-17A⁺CD4⁺, and IL-11⁺ IL-17A⁺CD4⁺ T cells were significantly increased in the CSF in comparison with the matched PBMCs, indicating the accumulation of those cells in the CSF (Fig. 1).

FIGURE 1.

IL-11⁺CD4⁺ and Th17 cells are enriched in the CSF from RRMS patients in comparison with the matched blood samples. (A) PBMCs and CSF cells were separated from blood and CSF samples from six untreated RRMS patients. The cells were stimulated with PMA and ionomycin for intracellular staining. The percentage of cells expressing the indicated cytokines was determined in gated CD4⁺ T cells. Each symbol represents an individual donor, and horizontal bars represent mean value. Statistical analysis was performed using a two-tailed paired Student t test. (B) Representative staining of the indicated cytokines in the gated CD4⁺ cells from one out of six untreated RRMS patients.

Immunohistochemistry studies of diagnostic brain MS lesion biopsy samples (from five additional patients whose demographic, brain MRI, and pathology data are presented in the Supplemental Table I) revealed an enrichment of IL-11⁺CD4⁺ cells within the infiltrating cells in both myelinated (average 47.9%) and demyelinated areas (average 62.8%) (Fig. 2) in comparison with the percentages of IL-11⁺CD4⁺ cells in peripheral circulation (Fig. 1A). The results suggest an early accumulation of IL-11⁺CD4⁺ cells in the active MS lesions.

FIGURE 2.

Immunohistochemistry studies of the active brain MS lesions reveal accumulation of IL-11⁺CD4⁺ cells in the inflammatory infiltrates. (A) Active brain MS lesion tissue from five patients obtained by biopsy was used for CD4, IL-11, and IL-17A staining. Presented is an average percentage of IL-11⁺ and IL-17A⁺ CD4⁺ cells ± SD in the myelinated and demyelinated tissue. (B) Representative staining from one out of five patients. Cell nuclei were visualized in DAPI channel (blue), CD4 in Cy3 channel (green), and IL-11 and IL-17A in Cy5 (red). Scale bars, 1 μm (H&E and LFB panels).

IL-11 induces the expansion of IL-17A– and CCR6-expressing CD4⁺ T cells

To examine the causative effect of IL-11 in the induction of Th17 cell responses and on their migration to the CNS, we have extended our studies to RREAE in SJL/J mice, an established animal model of the disease. Studies of the spleen cells from nine naive SJL/J mice revealed that IL-11 in vitro stimulation induces a significant expansion of IL-17A⁺CD4⁺ and CCR6⁺CD4⁺ cells (Fig. 3A). IL-11 induced STAT3 phosphorylation and RORγτ expression in a dose-dependent fashion, which was reversed in the presence of αIL-11 mAb (Fig. 3B).

FIGURE 3.

IL-11 induces STAT3 phosphorylation and RORγτ expression. (A) Spleen cells from nine naive SJL/J mice were cultured in the absence or presence of IL-11 (100 ng/ml), followed by intracellular staining for IL-17A and surface staining for CCR6. The percentage of cells expressing each molecule was determined in gated CD4⁺ T cells. Each symbol represents one mouse; horizontal bars represent mean values. Statistical analysis was performed using a two-tailed paired Student t test. (B) Spleen cells were separated from three naive SJL/J mice and incubated in the absence or presence of IL-11 (10, 50, 100 ng/ml) and αIL-11 mAb. Cell lysates were prepared following 30 min of culture and used for Western blotting to detect total and pSTAT3, STAT3, RORγτ, and β-actin expression.

IL-11 induces exacerbation of RREAE and increased numbers of IL-17A⁺CD4⁺ cells in the spinal cord inflammatory infiltrates

The causal effect of IL-11 on the development of the Th17-mediated response was next examined in RREAE. Following immunization of SJL/J mice with PLP_139–151, the administration of recombinant mouse IL-11 for 10 consecutive days starting at the peak of the first relapse (day 12 postimmunization) induced an acute exacerbation and increase in mean clinical scores that remained elevated over the entire course of the disease (Fig. 4A). On day 70 postimmunization, the number of inflammatory foci was enumerated by a blinded neuropathologist and found to be significantly higher (2.9-fold) in the spinal cord parenchyma of mice that received IL-11 in comparison with control RREAE (Fig. 4B) (13).

FIGURE 4.

IL-11 induces worsening of the clinical disease course in RREAE. (A) Twelve female 8- to 12-wk-old SJL/J mice were immunized with 80 μg/mouse of PLP_139–151 peptide. Starting from day 12 postimmunization, six mice per group received i.p. injections of recombinant mouse IL-11 (0.5 μg/mouse) or control vehicle daily for 10 d. Clinical scores were assigned daily for 70 d. Presented are mean clinical scores ± SD. Statistical analysis was performed using a two-tailed Student t test. (B) On day 70 postimmunization, mice were sacrificed, and the brain and spinal columns were removed and fixed. The paraffin sections were stained with Luxol fast blue and H&E. Parenchymal inflammatory foci (>10 mononuclear cells) were quantified by a blinded neuropathologist, and results are presented as mean ± SD per group. Statistical analysis was performed using a two-tailed Student t test. (C) Representative sections demonstrate more meningeal and parenchymal inflammation and larger areas of demyelination in the IL-11–treated versus the control mouse spinal cord. Luxol fast blue and H&E stain. Scale bar, 50 μm.

To test whether IL-11 induces a Th17 response in the acute phase of disease, we injected a single dose of the recombinant mouse IL-11 (0.5 μg/mouse) or control vehicle (six mice per group) on day 12 postimmunization. After 16 h, the mice were sacrificed, and cells were harvested for further study. Flow cytometry studies revealed that IL-11 significantly increased the percentage of IL-17A⁺CD4⁺ cells in PBMCs and in the spinal cord inflammatory infiltrates (Fig. 5A, 5B), whereas it did not change in other niches. Th17 cells in both compartments expressed high levels of CCR6 (Fig. 5C). In contrast, the numbers of IFN-γ⁺ and IL-4⁺CD4⁺ cells did not change (data not shown). IL-11 also induced increased serum mean ± SD IL-17A levels in comparison with the control RREAE (Fig. 5D).

FIGURE 5.

IL-11 induces Th17 cell response in RREAE. (A) SJL/J mice were immunized with PLP_139–151 peptide. On day 12 postimmunization, six mice per group were i.p. injected with recombinant mouse IL-11 (0.5 μg/mouse) or vehicle control. After 16 h, the mice were sacrificed and tissues were harvested. The cells were stimulated with PMA and ionomycin for intracellular staining, and the percentage of cells expressing each molecule was determined in gated CD4⁺ T cells. (B) Representative staining. (C) The percentage of CCR6-expressing cells in gated IL-17A⁺CD4⁺ cells. (D) The concentration of IL-17A in serum samples was measured by ELISA. (E) On day 56 postimmunization, six mice per group received recombinant mouse IL-11 (0.5 μg/mouse) or control vehicle. After 16 h, the mice were sacrificed. The percentage of IL-17A⁺ cells was determined in gated CD4⁺ T cells from the PBMCs, spleen, and spinal cord inflammatory infiltrates. The percentage of ICAM-1⁺ cells was determined in gated CD4⁺ T cells from brain cell infiltrates. (F) The gene expression of IL-11 in LN and IL-17A in brain inflammatory infiltrates was detected by RT-PCR. (G) The concentration of IL-17A in the serum samples was measured by ELISA. (H) The percentages of ICAM-1⁺ cells were determined in gated IL-11⁺CD4⁺ T cells from the brain and spinal cord cell infiltrates. The percentage of CCR6⁺ cells was determined in IL-11⁺CD4⁺ T cells from the spinal cord infiltrates. The results are presented as mean ± SD per group. Statistical analysis was performed using a two-tailed paired Student t test. Each experiment was performed once.

We next studied the effect of IL-11 in the relapsing-remitting (RR) phase of the disease, which may be more relevant for human RRMS. A single dose of IL-11 was administered at day 56 postimmunization and again increased the percentage of IL-17⁺CD4⁺ cells in the PBMCs, spleen, and spinal cord infiltrates within 16 h (Fig. 5E). Studies of the adhesion molecules revealed that IL-11 administration during the RR phase of the disease increased ICAM-1 expression (mean ± SD) in brain-derived CD4⁺ cells (Fig. 5E). Gene expression studies at the same time point revealed increased IL-11 expression in the LN2s and increased IL-17A expression (mean ± SD) in the brain inflammatory infiltrates (Fig. 5F). Serum cytokine measurements at the same time revealed increased IL-17A levels following IL-11 administration, in comparison with control RREAE (Fig. 5G). Interestingly, IL-11 administration during the RR phase of the disease induced a significant increase in ICAM-1 expression on IL-11⁺CD4⁺ cells from the brain and spinal cord as well as CCR6 expression on spinal cord–derived IL-11⁺CD4⁺ cells (Fig. 5H), whereas the other tested markers did not change (data not shown).

Ex vivo signaling studies revealed that a single IL-11 administration during the RREAE induced a significantly increased phosphorylation of NF-κΒ p65 and the expression of transcription factor RORγτ (Supplemental Fig. 1). These results are consistent with our study in CIS patients, in which IL-11 induced in a dose-dependent manner NF-κΒ p65 phosphorylation and RORc expression, which were reversed by blocking αIL-11 mAb (Y. Tao, unpublished results).

IL-11 induced encephalitogenic CD4⁺ cells in an in vitro Ag recall assay

To test the hypothesis that IL-11 induces encephalitogenic CD4⁺ cells, we cultured spleen- and LN-derived cells from PLP_139–151-immunized SJL/J mice with PLP peptide, in the absence or presence of IL-11 and the Th17-polarizing cytokine IL-23. IL-11 induced a significant increase in the percentage of IL-17A⁺CD4⁺ cells, similar to IL-23, in comparison with the control Ag-stimulated cells. Ag recall in the presence of IL-11 also significantly increased the percentage of GM-CSF⁺, double-positive IL-17A⁺GM-CSF⁺, and IL-17A⁺IFN-γ⁺CD4⁺ cells as well as IL-21⁺CD4⁺ cells (Fig. 6A), suggestive of the encephalitogenic phenotype of IL-11–polarized Ag-reactive CD4⁺ cells. To further characterize the IL-11–induced CD4⁺ cell surface markers, we measured their expression of the activation markers CD44, CD28, and PD-1; adhesion molecules ICAM-1 and VLA-4; chemokine receptors CCR5 and CCR6; and cytokine receptors IL-21R, IL-23, and IL-6R. Our data demonstrate that IL-11 induced the expansion of IL-6R⁺CD4⁺ cells (Fig. 6A), which may reflect their responsiveness to IL-6, a Th17-promoting cytokine, whereas the other tested marker did not change (data not shown).

FIGURE 6.

IL-11 induced encephalitogenic CD4⁺ cells. Spleen and LN cells were separated from PLP_139–151-immunized SJL mice on day 11 postimmunization. Cells were cultured with PLP peptides (20 μg/ml) in the absence or presence of IL-11 (100 ng/ml) and the control Th-17–polarizing cytokine IL-23 (20 ng/ml) for 7 d. (A) Cells were restimulated with PMA/ionomycin for intracellular staining. The percentages of the cells expressing each molecule in gated CD4⁺ T cells (n = 12), (B) CD8⁺ cells (n = 6), and (C) CD19⁺ cells (n = 6) were determined by flow cytometry. Each symbol represents one mouse, and horizontal bars represent mean values. Statistical analysis was performed using repeated measures ANOVA. *p < 0.05, **p < 0.01, ***p < 0.001.

In addition, we measured the capacity of IL-11 to induce the expansion of CD8⁺ and CD19⁺ B cells. Studies of the CD8⁺ gated cells revealed the expansion of IL-17A⁺CD8⁺ cells (Fig. 6B), whereas their expression of IL-21, IL-23, VLA-4, and IL-6R was not significantly changed (data not shown).

In vitro studies of the CD19⁺ gated B cells demonstrated that IL-11 induced the expansion of TNF-α⁺ and IL-1α⁺CD19⁺ cells and decreased the numbers of IL-10⁺ B regulatory cells (Fig. 6C), whereas the percentages of GM-CSF⁺, IL-23⁺, VLA-4⁺, IL-6R⁺, and IL-21R⁺ B cells did not change in comparison with the control cultures.

Passive transfer of IL-11–induced encephalitogenic CD4⁺ cells induces RREAE

Passive transfer experiments were performed to detect whether IL-11–polarized Ag-specific CD4⁺ cells are able to induce RREAE in naive recipients. Separated CD4⁺ T cells from immunized mice that were cultured with PLP_139–151 in the absence or presence of IL-11 and control Th-17–polarizing cytokine IL-23 were injected i.p. in the naive mice. The mice that received IL-11 in vitro–polarized CD4⁺ cells had more-severe disease in comparison with the recipients of CD4⁺ cells stimulated with PLP_139–151 only (Fig. 7A). The difference in disease severity between the recipients of IL-11– and IL-23–polarized CD4⁺ cells was not statistically significant. On day 70 post–passive transfer, the inflammatory foci were enumerated by a blinded neuropathologist and found to be significantly higher (1.5-fold) in the spinal cord meninges of mice that received IL-11 in comparison with control RREAE (Fig. 7B).

FIGURE 7.

Passive transfer of IL-11–stimulated CD4⁺ cells induces severe RREAE. (A) CD4⁺ T cells were isolated from the cultures described in Fig. 6. A total of 20 × 10⁶ CD4⁺ T cells per recipient were injected into naive SJL/J mice. Clinical scores of the six recipient mice per group were assigned daily for 70 d. Statistical analysis was performed using a repeated measures ANOVA. (B) On day 70 post–passive transfer, mice were sacrificed, and the brain and spinal cords were stained with Luxol fast blue and H&E. Meningeal inflammatory foci (>10 mononuclear cells) were quantified by a blinded neuropathologist, and statistical analysis was performed using a two-tailed Student t test. (C and D) Single-cell solutions were obtained from recipient mice sacrificed at day 14 post–passive transfer and incubated with PMA/ionomycin for intracellular staining. The percentages of cells expressing indicated markers in gated CD4⁺ cells, (E–G) viable cells, and (H) gated CD19⁺ B cells were determined by flow cytometry. (I) In mice sacrificed at day 70 post–passive transfer, the percentage of cells expressing the indicated markers in the viable cells and (J) gated CD4⁺ cells was determined by flow cytometry. The results represent mean ± SD values per group. Statistical analysis was performed using paired t test. *p < 0.5.

In a repeated experiment, the recipient mice were sacrificed at the peak of the disease (day 14), and flow cytometry studies were performed on the PBMC, spleen, LN, and CNS-derived mononuclear cells. The results revealed that the recipients of IL-11–polarized CD4⁺ cells had a higher percentage of IL-17A⁺CD4⁺ cells in LN- and CNS-derived mononuclear cells, a higher percentage of CCR6⁺CD4⁺ cells in the CNS, and a higher percentage of IL-17A⁺CCR6⁺CD4⁺ cells in the LN and spleen cells in comparison with the control recipient mice (Fig. 7C), whereas these markers did not change in the other niches. The numbers of IL-11⁺, IFN-γ⁺, and IL-21⁺ CD4⁺ cells did not change (data not shown). Studies of the adhesion molecules revealed an increased percentage of VLA-4⁺CD4⁺ cells in the PBMC and spleen cells of the recipients of IL-11–polarized Ag-specific CD4⁺ cells (Fig. 7D), whereas the expression of ICAM-1 did not change (data not shown).

Interestingly, the recipients of the IL-11–polarized CD4⁺ cells also exhibited an expansion of neutrophils in the PBMC, spleen, and CNS-derived mononuclear cells (Fig. 7E), which was likely induced by IL-17A (15). In addition, transferring IL-11–polarized CD4⁺ cells caused the expansion of CD8a⁺ cells in LNs (Fig. 7F) as well as the expansion of CD19⁺ cells in the CNS infiltrates (Fig. 7G) and increased percentage of IL-21R⁺CD19⁺ cells in LNs (Fig. 7H). This may be related to the IL-11–induced IL-21 secretion by CD4 cells, as demonstrated in the Ag recall assays (Fig. 6A).

In mice sacrificed on day 70 following passive transfer, flow cytometry studies revealed a persisting increase of neutrophils in PBMCs (Fig. 7I) as well as an increased percentage of CCR6⁺CD4⁺ cells in the LNs of recipients of IL-11–polarized CD4⁺ cells, in comparison with the control recipient mice (Fig. 7J), whereas the other markers did not change.

IL-11 enhances severity of the disease induced by passive transfer of CNS-derived nonmanipulated mononuclear cells

Because our data from mice with PLP_139–151-induced RREAE show that the percentage of Th17 cells in the CNS is significantly higher than in the peripheral lymphoid organs, we proposed to passively transfer CNS-derived cells to detect whether they can induce the disease even without in vitro polarization and to see whether CNS-derived cells from immunized mice that received IL-11 have enhanced encephalitogenic potential.

A total of 2 × 10⁶ mononuclear cells were isolated from the brain and spinal cord tissue and transferred immediately to recipient naive mice through i.p. injection. The recipients of cells from donors who received a single IL-11 dose 16 h prior to cell harvesting had significantly higher mean ± SD clinical scores in comparison with the mice that received CNS-derived cells from the control RREAE donors (Fig. 8A). Passive transfer of CNS-derived mononuclear cells from mice that received IL-11 induced a higher percentage of IL-17A⁺ and CCR6⁺CD4⁺ cells in the recipients’ CNS in comparison with the control group (Fig. 8B).

FIGURE 8.

IL-11 enhances the disease severity induced by passive transfer of nonmanipulated CNS mononuclear cells from mice with RREAE. (A) SJL/J mice that were immunized with PLP_139–151 received either IL-11 (0.5 μg) or control 1× PBS injection on day 11 postimmunization. The mice were sacrificed after 16 h. A total of 2 × 10⁶ mononuclear cells from the CNS infiltrate were isolated using a Percoll gradient and transferred to recipient mice via i.p. injection. Clinical scores of six recipient mice per group were assigned daily for 70 d. (B) On day 70 after passive transfer, CNS-infiltrating mononuclear cells of the recipient mice were isolated by Percoll gradient. Cells were restimulated with PMA/ionomycin for intracellular staining. The percentages of CD4⁺ cells expressing each molecule were determined by flow cytometry. The results represent mean ± SD values. Statistical analysis was performed using repeated measures paired t test. Each experiment was performed once. *p < 0.05.

Discussion

We have previously reported that IL-11 and IL-17A CSF and serum levels are significantly increased in the CIS and RRMS patients in comparison with the HCs. Serum levels of both cytokines are increased during the clinical relapses in comparison with the remissions of untreated RRMS patients (2). In the current study, we found that IL-11⁺CD4⁺ and IL-17A⁺CD4⁺ T cells accumulate in the CSF of RRMS patients and in the active brain MS lesions, suggesting their role in the lesion formation. Following the previously reported role of IL-11 in the in vitro differentiation and expansion of Th17 cells in early MS, our new findings suggest that IL-11 may induce Th17 cell responses in vivo, which may constitute an initial step in the development of the autoimmune response in MS.

Trafficking of the activated myelin-reactive T cells across the blood–brain barrier (BBB) into the CNS is a central event in the pathogenesis of MS (16–19). The migration of the inflammatory cells through the choroid plexus and BBB is regulated by multiple adhesion molecules and chemokine receptors. In the RREAE animal model of the disease, CCR6⁺ Th17 cells accumulate in the CNS, whereas passive transfer of CCR6-deficient CD4⁺ cells did not induce a disease, supporting a critical role of CCR6 in the early migration of inflammatory cells to the CNS (3). In vitro animal studies reveal that IL-11 induces IL-17A and CCR6 expression in CD4⁺ cells. In the same experiments, IL-11 induced STAT3 phosphorylation in a dose-dependent manner. Phosphorylated STAT3 is a transcription factor that mediates IL-6–induced Th17 cell differentiation and IL-23–mediated Th17 cell expansion (20). Furthermore, NF-κΒ c-Rel and RelA/p65 transcriptionally activate RORγτ (21), which induces CCR6 expression in mouse CD4⁺ cells (22). We therefore propose that the IL-11–mediated induction of CCR6 expression may be regulated via RORc induced by STAT3 and p65 phosphorylation (21, 22).

IL-11 has been implicated in embryonal implantation (23) and metastatic cancer cell migration through IL-11R–induced NF-κΒ signaling (24). However, its role in the migration of inflammatory cells across the BBB has not been examined. Our results indicate that administration of IL-11 at the time of clinical onset of RREAE induced an acute exacerbation and increased clinical scores, which persisted during the entire course of the disease. IL-11 increased the numbers of spinal cord inflammatory foci, the numbers of peripheral and CNS-infiltrating IL-17⁺CD4⁺ cells, and IL-17A serum levels. In addition, IL-11 increased the numbers of ICAM-1⁺CD4⁺ T cells in the RREAE brain inflammatory infiltrates (2). ICAM-1 is an NF-κΒ–regulated adhesion molecule involved in the trans-endothelial migration to the CNS (25). ICAM-1–deficient mice showed attenuated EAE and dramatically reduced spinal cord T cell infiltrates. Furthermore, passive transfer of Ag-restimulated ICAM-1⁻/⁻ T cells failed to induce EAE in wild-type mice (26). Finally, IL-11 mediated increased expression of CCR6 and ICAM-1 on IL-11⁺CD4⁺ T cells from CNS infiltrates, indicating that IL-11–induced CCR6 and ICAM-1 may mediate IL-11⁺CD4⁺ cell migration to the CNS.

IL-6 and IL-11 induce inflammatory cell migration by regulating the expression of adhesion molecules and chemokine receptors. IL-6 induces VLA-4 expression on human CD4⁺CD45RO⁺ cells, and IL-11 directly mediates the in vitro migration of CD4⁺ cells from HCs (27). IL-11 is reported to induce VLA-2 and ICAM-1 expression in endometrial epithelial cells (28); however, its effect on brain endothelial cells (ECs) has not been examined. Finally, Th17 cell migration across the BBB-EC monolayer via IL-17R and IL-22R signaling is mediated via the disruption of tight junction proteins (29). Both IL-17A (30) and IL-22 (31) induce IL-11 secretion, and our previous results identified a positive correlation between IL-22 and IL-11 levels in the CSF of CIS patients (2). IL-17 also induces IL-6 and neutrophil-attracting chemokine CXCL8 (IL-8) expression by BBB ECs (32), which may enhance both neutrophil and CD4⁺ cell migration through the BBB. Our results indicate that, similar to IL-6, IL-11 induces CCR6 expression, which has been reported to mediate IL-17A⁺CD4⁺ cell migration through the choroid plexus (3).

We have demonstrated that IL-11 induces encephalitogenic CD4⁺ cells, which induce RREAE upon passive transfer to naive recipients. In Ag recall assays, IL-11 induced a significant expansion of IL-17A⁺, GM-CSF⁺, and IL-21⁺CD4⁺ cells. As stated in our previously published paper (2), IL-11 did not increase the expression of FOXP3 or the secretion of TGF-β or IL-10 in FOXP3 memory CD4⁺ T cells. Because we did not detect IL-11–mediated induction of T regulatory cells in CIS patients, we did not pursue those studies in mice. Notably, a recent study (33) reported that IL-11 in Ag recall experiments from MBP-immunized mice induced Foxp3 gene expression, warranting future studies of the effect of IL-11 on T regulatory cells. Interestingly, IL-11 also induced the expansion of IL-17⁺CD8⁺ cells and increased the percentages of proinflammatory TNF-α⁺ and IL-1α⁺B cells, whereas it decreased the numbers of regulatory IL-10⁺B cells. Passive transfer of IL-11–polarized Ag-specific CD4⁺ cells into naive recipients induced severe RREAE, comparable to the prototypical encephalitogenic IL-23–polarized CD4⁺ cells. Further characterization of the mechanisms involved in IL-11–polarized CD4⁺ cell–induced RREAE revealed increased numbers of IL-17A⁺ and CCR6⁺CD4⁺ cells in both peripheral lymphoid organs and the CNS and increased numbers of VLA-4⁺CD4⁺ cells, which preferentially migrate to the CNS. Interestingly, transfer of IL-11–polarized CD4⁺ cells also induces increased numbers of neutrophils in the periphery and the CNS, both acutely and in the chronic phase of the disease, suggesting that IL-11 may contribute (similar to the IL-23/IL-17A/GM-CSF axis) to the regulation of neutrophil numbers in the peripheral circulation (34). We propose that IL-11–induced expansion of neutrophils may constitute a key event in the IL-11–induced clinical worsening because the increased numbers of neutrophils precede EAE clinical symptoms and neutrophil blockade prevents the clinical onset of disease (15). The role of neutrophils in the opening of the BBB has been well established, as even their adhesion to the endothelial barrier may impair intraendothelial contacts via secretion of proteases and free oxygen radicals. However, this study reports an IL-11–induced neutrophil expansion.

Further characterization of the inflammatory responses induced by IL-11–stimulated CD4⁺ cells also revealed expansion of B cells within the recipients’ CNS. It is tempting to speculate that a proportion of IL-11–induced IL-21⁺CD4⁺ cells may represent T follicular helper cells, which induce B cell activation and autoantibody production.

As a corollary, passive transfer of the nonmanipulated CNS-infiltrating cells from mice with RREAE that received one dose of IL-11 induced more-severe disease and increased numbers of IL-17A⁺ and CCR6⁺CD4⁺ cells in the CNS of recipient mice than the transfer of control RREAE CNS cells. The findings are suggestive of their encephalitogenic potential even in the absence of Ag restimulation and cytokine polarization prior to the transfer.

Disclosures

The authors have no financial conflicts of interest.