Lack of adiponectin leads to increased lymphocyte activation and increased disease severity in a mouse model of multiple sclerosis

Abstract

Multiple sclerosis (MS) is a presumed autoimmune disease directed against central nervous system (CNS) myelin, in which diet and obesity are implicated as risk factors. Immune responses can be influenced by molecules produced by fat cells, called adipokines. Adiponectin is an adipokine with anti-inflammatory effects. We tested the hypothesis that adiponectin has a protective role in the EAE model for MS, that can be induced by immunization with myelin antigens or transfer of myelin-specific T lymphocytes. Adiponectin deficient (ADPKO) mice developed worse EAE with greater CNS inflammation, demyelination, and axon injury. Lymphocytes from myelin-immunized ADPKO mice proliferated more, produced higher amounts of IFN-γ, IL-17, TNF-α, IL-6, and transferred more severe EAE than wild type (WT) lymphocytes. At EAE peak, the spleen and CNS of ADPKO had fewer regulatory T (Treg) cells than WT mice and during EAE recovery, Foxp3, IL-10 and TGF-β expression levels in the CNS were reduced in ADPKO compared with WT mice. Treatment with globular adiponectin in vivo ameliorated EAE, and was associated with an increase in Treg cells. These data indicate that adiponectin is an important regulator of T-cell functions during EAE, suggesting a new avenue of investigation for MS treatment.

Keywords: Adiponectin; EAE; Immunomodulation; Multiple sclerosis.

Conflict of interest statement

The authors have declared that no financial conflict of interest exists for the subject of this article.

© 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

Figures

Figure 1. Exacerbation of EAE in ADPKO mice

(A) Clinical course of EAE induced in ADPKO (n=6) and WT mice (n=7) with MOG35-55. ADPKO mice displayed significantly worse clinical EAE than WT control mice (P<0.0001 by two-way ANOVA). This is one representative experiment out of seven performed. Data are mean clinical score ± SEM. (B-E) Inflammation and demyelination were assessed using H&E and Luxol fast blue staining of spinal cord sections from immunized WT (B,D) (n=5) and ADPKO mice (C,E) (n=7) at disease peak. Inflammatory cells (black arrows) were mainly localized in the meninges and less so in the white matter in WT mice, while they extensively infiltrated the parenchyma in the ADPKO mice. (F,G) Spinal cord sections from EAE affected WT and ADPKO mice were stained for myelin basic protein (in red) and SMI-32 (in green), to detect myelin and damaged axons, respectively. White arrows show damaged axons. Original magnification, 4X (B-E); 63X (F,G). Scale bars, 100 μm (B-E); 10 μm (F,G). (H) Quantification of inflammation, demyelination and axonal damage. Horizontal lines are median values. (I) Quantitative real-time PCR studies of spinal cord mRNA at EAE peak (n=5/group) for the expression of the T cell marker CD3, adhesion molecules VCAM-1, ICAM-1 and pro-inflammatory cytokines IL-17A, IFNγ, TNFα, and IL-6. Quantification is relative to expression levels in spinal cords from naïve mice. Data are representative of two experiments performed. (J) MOG35-55 specific IFNγ (Th1) and IL-2 producing cells in the CNS of immunized ADPKO and WT mice determined at disease peak by ELISPOT (n=10/group). Data are presented as absolute number of cytokine producing cells per CNS and they are representative of two experiments performed. NS: not significant.

Figure 2. Enhanced T cell proliferation and cytokine production in response to MOG35-55 in ADPKO compared to WT mice

(A-B) Lymph node cells were isolated from immunized ADPKO and WT mice on day 12 pi and then restimulated in vitro with MOG35-55 or medium alone. (A) Lymph node cell proliferative response to MOG35-55 in ADPKO and WT mice (n=5/group). Data are mean CPM ± SD, and they are representative of four experiments performed. (B) IFNγ, IL-17, TNFα, and IL-6 concentrations were measured by cytometric bead array (CBA), and leptin was measured by ELISA in supernatants of cultured lymph node cells from MOG35-55 immunized ADPKO and WT mice from day 12 pi (n=5 per group). Results shown as means ± SD. Data are representative of three different experiments. CPM=counts per minute.

Figure 3. MOG35-55 specific Th1 cells from ADPKO mice induce more severe EAE than WT Th1 cells

(A) Cytokine production analyzed by intracellular staining by flow cytometry for INFγ TNFα and IL-17 of Th1 cells obtained from ADPKO and WT mice. The numbers in the quadrants represent the percentages of cytokine-producing cells. Two separate experiments were performed with similar results. (B) Clinical course of adoptively transferred EAE induced by injection of activated MOG35-55 -specific Th1 cells from ADPKO or WT donors. Nine naïve C57BL/6 WT recipient mice received ADPKO Th1 cells, and 8 WT recipient mice received WT Th1 cells. MOG35-55 -specific Th1 cells from ADPKO donor mice induced a significantly more severe disease compared to Th1 cells from WT mice (P<0.001 by two-way ANOVA). This experiment is representative of two performed with similar results. (C) Clinical course of adoptively transferred EAE into ADPKO (n=8) versus WT (n=9) recipient mice, induced by injection of activated MOG35-55 -specific CD4+ T cells obtained from C57BL/6 WT mice. The shown experiment is representative of two performed with similar results. In B and C, graphs represent mean clinical score ± SEM of all animals in each group.

Figure 4. ADPKO APCs do not differ from WT APCs in supporting proliferation of CD4+ T cells

ADPKO or WT CD4+ T cells purified from draining lymph nodes from immunized ADPKO and WT mice on day 11 pi were cultured, with or without MOG35-55, in presence of irradiated CD4- cells from either ADPKO or WT mice as APC. No significant differences were noted in CD4+ T cell proliferation (A) or cytokine production (B) in the presence of ADPKO APCs compared to WT APCs. Significant differences were present when comparing proliferation and IL-17, IFNγ and TNFα production by ADPKO CD4+ T cells versus WT CD4+ T cells regardless of APC type. Results are means ± SD. Data are representative of two different experiments performed.

Figure 5. Defect in Treg cells in the spleens and CNS of ADPKO mice during EAE

(A) Splenocytes and (B) CNS mononuclear cells from immunized WT and ADPKO mice at EAE peak (n=10/group). Tregs were evaluated by flow cytometry, representative plots are in the top panels. Lower panel bar graphs represent the mean values of all mice analyzed. Percentages of CD25+ Foxp3+ cells within the CD4+ cell population were calculated. Absolute numbers of CD4+CD25+Foxp3+ Treg cells were calculated based on the total number of splenocytes or CNS mononuclear cells isolated/mouse. Results are means ± SD. (C) mRNA was isolated from spinal cords isolated at EAE peak (day 16-18 pi) and (D) during the EAE recovery phase (day 30 pi) from ADPKO and WT mice (n=5/group). Expression of Treg marker Foxp3, and of anti-inflammatory cytokines IL-10 and TGF-β was studied by quantitative RT-PCR. Data are mean ± SD and representative of two experiments performed. Quantification is relative to expression levels in spinal cords from naïve mice. NS: not significant.

Figure 6. Defective suppressive function of ADPKO Tregs

(A) CD4+CD25+ Treg cells and CD4 +CD25- effector T cell were isolated from spleen and peripheral lymph nodes of naïve WT and ADP KO mice. T effector cell proliferation was induced with an anti-CD3 antibody. CD4 +CD25- effector T cell from WT mice and WT or ADPKO CD4+CD25+ Treg cells were cocultured at different ratios as indicated (no Treg, 2:1 and 4:1) with CD4- irradiated APCs. T effector cell were labeled with CFSE and proliferation measured by CFSE dilution (numbers in the plots are percentages of proliferating cells). This experiment was performed twice with similar results. In (B) the results of both experiments are expressed as percentage of inhibition of T effector cell proliferation by either WT or ADPKO T reg cells, with proliferation in the absence of T regs being 100%. (C) AdipoR1 and R2 expression by Western blot of CD4+CD25+ Treg cells isolated from naïve and MOG 35-55 immunized mice (lanes 1 and 2 respectively), and of T effector cells isolated from naïve mice (lane 3) and naïve splenocytes (lane 4).

Figure 7. Treatment with gADP is protective in EAE and increases the number of Treg cells

EAE was induced in C57BL/6 WT mice with MOG35-55. Treatment with gADP (n=5) or PBS (n=10) was initiated on day 5 pi. Mice received either 25 μg gADP/mouse or same volume of PBS intraperitoneally twice daily until day 12 pi. (A) EAE clinical course in the gADP and PBS treated groups. Neurological signs were assessed on a scale from 0 to 5. Data are mean clinical score ± SEM. There was 80% incidence of EAE (8/10) in the PBS treated group vs. 20% (1/5) in the gADP treated group. The single EAE-affected mouse in the gADP group reached a score of 2. For the control mice, mean maximal clinical score for this experiment was 3 ± 0.5, including the two that didn’t develop EAE. The experiment was terminated on day 15 pi when four of PBS-treated control mice had reached clinical score of 4 and one had died. This experiment was performed twice with similar results. (B) mRNA expression levels of the T cell marker CD3, the cytokines IFNγ, IL-17, TNFα and IL-6, and the adhesion molecules ICAM-1 and VCAM-1 evaluated by RT-PCR in spinal cords obtained on day 15 pi (gADP, n=3 and PBS, n=5). Quantification is relative to expression levels in spinal cords from naïve mice. (C-D) Percentages and absolute numbers of Treg cells were studied by flow cytometry in draining lymph nodes and spleens of gADP and PBS treated mice (day 15 pi). CD25+ Foxp3+ cells percentages were calculated within the gated CD4+ cells. Absolute numbers were calculated based on the number of lymph node cells/mg of tissue or total splenocytes/mouse. The bar graphs represent the results compiled from gADP (n=5) and PBS (n=10) treated mice. Results are means ± SD.