Exercise training attenuates experimental autoimmune encephalomyelitis by peripheral immunomodulation rather than direct neuroprotection

Experimental Neurology 299 (2018) 56–64

Contents lists available at ScienceDirect

Experimental Neurology journal homepage: www.elsevier.com/locate/yexnr

Research paper

Exercise training attenuates experimental autoimmune encephalomyelitis by peripheral immunomodulation rather than direct neuroprotection

MARK

Ofira Einsteina,⁎, Nina Fainsteinb, Olga Touloumic, Roza Lagoudakic, Ester Hanyaa, Nikolaos Grigoriadisc, Abram Katza, Tamir Ben-Hurb a b c

Physical Therapy Department, Faculty of Health Sciences, Ariel University, Ariel, Israel Department of Neurology, The Agnes Ginges Center for Human Neurogenetics, Hadassah – Hebrew University Medical Center, Jerusalem, Israel B' Department of Neurology, AHEPA University Hospital of Thessaloniki, Greece

A R T I C L E I N F O

A B S T R A C T

Keywords: Exercise training Multiple sclerosis Experimental autoimmune encephalomyelitis Immunomodulation

Background: Conflicting results exist on the effects of exercise training (ET) on Experimental Autoimmune Encephalomyelitis (EAE), nor is it known how exercise impacts on disease progression. Objective: We examined whether ET ameliorates the development of EAE by modulating the systemic immune system or exerting direct neuroprotective effects on the CNS. Methods: Healthy mice were subjected to 6 weeks of motorized treadmill running. The Proteolipid protein (PLP)induced transfer EAE model in mice was utilized. To assess effects of ET on systemic autoimmunity, lymph-node (LN)-T cells from trained- vs. sedentary donor mice were transferred to naïve recipients. To assess direct neuroprotective effects of ET, PLP-reactive LN-T cells were transferred into recipient mice that were trained prior to EAE transfer or to sedentary mice. EAE severity was assessed in vivo and the characteristics of encephalitogenic LN-T cells derived from PLP-immunized mice were evaluated in vitro. Results: LN-T cells obtained from trained mice induced an attenuated clinical and pathological EAE in recipient mice vs. cells derived from sedentary animals. Training inhibited the activation, proliferation and cytokine gene expression of PLP-reactive T cells in response to CNS-derived autoantigen, but strongly enhanced their proliferation in response to Concanavalin A, a non-specific stimulus. However, there was no difference in EAE severity when autoreactive encephalitogenic T cells were transferred to trained vs. sedentary recipient mice. Conclusion: ET inhibits immune system responses to an auto-antigen to attenuate EAE, rather than generally suppressing the immune system, but does not induce a direct neuro-protective effect against EAE.

1. Introduction Multiple sclerosis (MS) is an immune-mediated disease of the central nervous system (CNS), leading to CNS demyelination and axonal damage (Frohman et al. 2006). During the course of MS, autoreactive T cells are activated in the peripheral lymphoid organs and migrate across the blood-brain barrier to induce inflammatory lesions within the CNS (McFarland and Martin 2007). MS causes chronic irreversible functional impairments and there is a requirement for therapeutic strategies that can reduce the deleterious impact of the disease on the quality of life of MS patients. Various clinical trials demonstrated the safety of exercise training (ET) in MS patients and suggested it may induce various physiological and functional beneficial effects (Motl and Pilutti 2012; Pilutti et al. 2014). Furthermore, there is growing evidence linking between ET and the immune system (Gjevestad et al. 2015; Gleeson et al. 2011; Walsh ⁎

et al. 2011). Regular physical activity reduces the risk of chronic diseases, partly owing to the anti-inflammatory effects of exercise (Baek 2016; Gleeson et al. 2011; Pruimboom et al. 2015; Spielman et al. 2016). Anti-inflammatory effects of ET are also documented in MS patients (Florindo 2014). These studies prompted additional research of ET for MS and specifically investigation of the beneficial effects and mechanisms of action of ET on Experimental Autoimmune Encephalomyelitis (EAE). EAE is the most commonly used animal model to study MS therapies (Robinson et al. 2014). EAE is characterized by T-cell and monocyte infiltration in the CNS, targeting proteins that are expressed by myelinproducing oligodendrocytes, such as proteolipid protein (PLP), myelin basic protein (MBP) and myelin oligodendrocyte glycoprotein (MOG). This inflammatory process results in demyelination, axonal damage and subsequently progressive hind limb paralysis (Kuerten and Angelov 2008; Robinson et al. 2014; Wekerle 2008).

Corresponding author at: Physiotherapy Department, Faculty of Health Sciences, Ariel University, Ariel 40700, Israel. E-mail address: ofi[email protected] (O. Einstein).

http://dx.doi.org/10.1016/j.expneurol.2017.10.008 Received 9 August 2017; Received in revised form 21 September 2017; Accepted 10 October 2017 Available online 12 October 2017 0014-4886/ © 2017 Elsevier Inc. All rights reserved.

Experimental Neurology 299 (2018) 56–64

O. Einstein et al.

for neurological symptoms up to 50 days post transfer.

Previous studies examined the effects of ET on EAE with conflicting results (Benson et al. 2015; Bernardes et al. 2016; Bernardes et al. 2013; Klaren et al. 2016; Patel and White 2013; Pryor et al. 2015; Rossi et al. 2009; Souza et al. 2016). It is not known whether ET mediates beneficial effects in EAE via modulation of the systemic immune system and/ or by inducing direct CNS neuroprotective effects. Therefore, the aims of the current study were: (1) to examine the clinical and neuropathological effects of ET in a clinically relevant model of MS; and (2) to investigate whether ET ameliorates EAE by modulating the systemic immune system, or exerting direct neuroprotective effects on the CNS in EAE. We employed here a unique experimental paradigm using the transfer EAE model that enabled us to distinguish between the potential systemic and central effects of ET in EAE. The data demonstrate that ET modulates the peripheral immune system responses to the myelin antigen to attenuate EAE, rather than inducing a CNS neuroprotective effect.

2.3. Forced treadmill exercise training (ET) Physical training and performance tests were performed on a 5-lane treadmill designed for mice (Panlab Harvard Apparatus, USA). The back of each treadmill lane contained an electrified grid, which delivered a shock stimulus to stationary mice (0.1 mA). Before commencement of exercise performance tests and training program, mice were familiarized with treadmill running for 10 min on three consecutive days (first day at 8 cm/s with no electrical shock; second and third days at 8 cm/s with 0.1 mA electrical shocks). Performance tests were performed prior to the training and at the end of a 6-week training protocol. There was a 72 h interval between the two types of performance tests. 2.3.1. Exhaustion speed performance test Maximal running speed for each mouse was assessed first by running the mice for 8 cm/s and then increasing the speed by 2 cm/s per minute until exhaustion (modified from (Qi et al. 2011)). Exhaustion was defined by an inability/refusal to continue when encouraged with a bottlebrush or a small puff of air.

2. Materials and methods 2.1. Experimental animals Female SJL/JCrHsd mice (6–7 weeks of age) were obtained from Envigo Inc., Israel. Animals were housed in communal cages at 22 ± 1 °C under a 12-h light/dark cycle (lights on at 07:00 hours), with free access to food and water. Animal experimentation has received approval by the Institutional Animal Care and Use Committee, and the studies were conducted in accordance with the United States Public Health Service's Policy on Humane Care and Use of Laboratory Animals. Behavioral experiments were performed between 8:00 a.m. and 4:00 p.m.

2.3.2. Exercise tolerance performance test Exercise tolerance was determined by running each mouse individually to exhaustion at 30 cm/s on a rodent-specific treadmill (modified from (Ritchie et al. 2014)). Exhaustion was defined as above. 2.3.3. Exercise training protocol Mice were subjected to a 6-week treadmill running, 5 days per week, 1 session per day at 23 cm/s. According to the baseline exercise speed performance tests, this training speed corresponds to an exercise intensity of 55–60% of maximal speed (Table 1). The trained mice were subjected to an incremental exercise training protocol. Each training session consisted of a 5-min warm-up at 8 cm/s. For the first week, warm-up was followed by 10 min of training, and for the second week by 20 min of training. During the following 4 weeks, warm-up was followed by 30 min training at 23 s/min. To minimize potential confounding factors such as differences in stress, sound and light exposure, sedentary control mice were left on the treadmill without running for the same duration as the exercise groups. No animal was excluded from the experiments.

2.2. Experimental design The transfer-EAE experimental set-up was designed to enable the differentiation of the effects of ET on systemic autoimmunity, specifically on induction of lymph node (LN)-derived T cell encephalitogenicity (part 1, Fig. 1A), vs. direct neuro-protective effects of ET on the CNS (part 2, Fig. 1B). Part 1 (Fig. 1A): To assess the modulatory effects of ET on systemic autoimmunity, we examined in vivo and in vitro the amount, potency and encephalitogenicity of LN-derived T cells from mice that underwent the ET program prior to PLP immunization (trained mice), compared with T cells from sedentary mice. To that end, healthy mice were subjected to a defined treadmill running program. This was followed by their immunization with a PLP peptide. Then, their LN-T cells were removed, stimulated in culture with PLP peptide and injected to naïve recipient mice, which developed EAE. Another group of recipient mice were injected with PLP-reactive LN T cells from sedentary mice and served as controls. Here we examined whether treadmill running of the donor mice modulated the systemic autoimmune process. Encephalitogenicity was examined (1) in vivo by examination of the clinical and pathological severity of EAE induced in recipient naïve SJL mice, following transfer of LN-T cells from trained- vs. sedentary donor mice; and (2) in vitro, at the day of LNCs removal or following secondary activation in vitro by the PLP auto-antigen, using activation and proliferation assays, as well as by T cell surface markers analysis and cytokine gene expression. Part 2 (Fig. 1B): Healthy mice were subjected to a defined treadmill running program (trained mice), followed by injection of PLP-reactive, encephalitogenic LN-T cells from donor mice. Sedentary mice were injected with the same PLP-reactive LN-T cells and served as controls. Here we examined whether treadmill running program of the recipient mice prior to transfer of encephalitogenic T cells attenuated the severity of EAE via direct neuro-protective effects on the CNS. EAE mice from both experimental protocols started to develop EAE clinical signs at 7–10 days post LN-T cell transfer, and were scored daily

2.4. Muscle enzyme activities At the end of the 6-week training period, trained and sedentary mice were sacrificed by decapitation 24 h after the last session (to minimize any potential residual effects of the last exercise bout). Soleus (type I, oxidative) and extensor digitorum longus (EDL; type II, glycolytic) muscles from trained and sedentary mice were dissected free, frozen in liquid nitrogen and analyzed for mitochondrial enzyme activities to assess the efficacy of the exercise protocol. Frozen muscles were freezedried, cleaned of non-muscle constituents, homogenized with ground glass homogenizers in ice-cold buffer (80 μl/mg dry wt) consisting of (in mM): 50 Tris-HCl, 1 EDTA, 0.05% (v/v) Triton X-100, pH 7.5. The extracts were centrifuged at 1400 × g (4 °C) for 1 min and the supernatant was assayed for citrate synthase (CS) and β-hydroxyacyl-CoA dehydrogenase (HAD) with standard spectrophotometric techniques, as described elsewhere (Zhang et al. 2007), as well as for protein (Biorad assay). Assays were conducted at 25 °C under conditions that yielded linearity with respect to time and extract volume. 2.5. Transfer experimental autoimmune encephalomyelitis (EAE) Proteolipid protein (PLP)139–151 transfer EAE model in 6–7- or 12–13 week-old female SJL/JCrHsd mice was utilized as previously 57

Experimental Neurology 299 (2018) 56–64

O. Einstein et al.

Fig. 1. Experimental outline. Transfer experimental autoimmune encephalomyelitis (EAE) model in mice was used to differentiate between effects of exercise training (ET) on systemic autoimmunity (A) and on the CNS (B). A: healthy mice were subjected to a 6 week-ET treadmill running program. At the end of the 5th week of training, trained mice were immunized with a PLP peptide and at the end of the 6th week their lymph node (LNs) were removed, stimulated for 72 h in culture with PLP peptide and injected to naïve recipient mice, which developed EAE. Another group of recipient mice was injected with PLP-reactive LN-T cells from sedentary mice and served as controls. PLP stimulated LN-T cells from exercised and sedentary mice were also analyzed in vitro for their activation and proliferation properties. In some experiments mice were sacrificed for histopathological analyses at 14 days post transfer, at the peak of the acute phase of the disease. B: Healthy mice were subjected to a defined treadmill running program and served as recipient mice to further develop EAE. Another group of naïve donor mice were immunized with PLP peptide and after 10 days their LN-T cells were removed, stimulated for 72 h in culture with PLP peptide and injected in either exercised or sedentary control recipient mice, which developed EAE. At both experimental set-ups mice developed clinical signs of EAE 7–10 days post transfer of LN-T cells and were scored daily for neurological symptoms up to 50 days post transfer.

Test type

Pre exercise (n = 10)

Post exercise (n = 10)

or in naïve donor mice prior to their immunization with PLP139–151 (Fig. 1). Recipient mice developed EAE and were scored daily for neurological symptoms up to 50 days after EAE induction as follows: 0 asymptomatic; 1 - partial loss of tail tonicity; 2 - atonic tail; 3 -hind leg weakness, difficulty to roll over, or both; 4 - hind leg paralysis; 5 - four leg paralysis; 6 - death due to EAE.

Exhaustion Speed (cm/s) Exercise Tolerance (min:s)

40 ± 1 14:18 ± 0:32

45 ± 1⁎ 22:33 ± 1:29⁎⁎⁎

2.6. Histopathological analyses

Table 1 Exercise training (ET) improves performance and citrate synthase activity in soleus muscle of mice. Performance tests

At day 14 after LN-T cell transfer (peak of EAE) animals were anesthetized with a lethal dose of sodium pentobarbital and subjected to perfusion via the ascending aorta with ice cold phosphate-buffered saline followed by 4% paraformaldehyde. The tissues were dissected and post-fixed in 4% paraformaldehyde for 24 h. Serial 6 μm paraffinembedded adjacent transverse sections were obtained from mid-cervical, mid-thoracic and mid-lumbar levels of the spinal cord. For histochemical staining, sections were stained with hematoxylin and eosin (H & E), Luxol fast blue (LFB)/nuclear fast red staining and Bielschowsky silver impregnation, to assess inflammation, demyelination, and axonal pathology, respectively, as previously described (Einstein et al. 2007). Immunohistochemistry was performed in adjacent serial sections with antibodies against macrophages (rat antimouse Mac3, 553,322, 1:800, BD Pharmingen), T cells (monoclonal rabbit anti–CD3, RM-9107-SO; 1:800, Thermo Scientific) and amyloid precursor protein (APP; monoclonal mouse anti-APP, MAB 348; 1:2000; Millipore). Bound antibody was visualized using an avidin-biotin technique. For each staining, three sections per mouse were quantified, one section per each spinal cord level. In order to ensure standardization and repeatability, the same researcher performed the embedding and the sectioning of all samples. Quantifications were performed spanning the whole white matter of the spinal cord stained sections. The number of immune cells in perivascular infiltrates was counted in H & E stained sections, and represented as total average number per square millimeter. Mac3 + and CD3 + cells were counted both in the

Muscle enzyme activities Enzyme

Muscle

Sedentary (n = 5)

Exercised (n = 4)

CS

Soleus EDL Soleus EDL

108 ± 9 133 ± 3 155 ± 6 89 ± 1

140 ± 4⁎ 121 ± 10 145 ± 2 94 ± 0

HAD

Citrate Synthase (CS) and Hydroxyacyl-Coenzyme A dehydrogenase (HAD) are given in nmol/min/mg dry wt. EDL – extensor digitorum longus. Data are represented as mean ± SE. ⁎ p < 0.05. ⁎⁎⁎ p < 0.001.

described (Einstein et al. 2007). Donor mice were immunized with 150 μg PLP139–151 peptide in 100 μl saline and an equal volume of complete Freund's adjuvant, containing 5 mg/ml H37RA (Difco Laboratories, Detroit, MI). Ten days after immunization, lymph nodes (LNs) were harvested and cultured in vitro for 72 h by seeding 3 × 106 cells/ml in Roswell Park Memorial Institute medium (RPMI) medium supplemented with 2.5% fetal calf serum, 1 mM L-glutamine, β-mercaptoethanol, and antibiotics in the presence of 100 μg/ml PLP139–151 peptide. Cells were then washed and 15 × 106 cells were injected intraperitoneally into recipient mice which were followed clinically. ET was conducted either in naïve recipient mice prior to LN-T cell transfer, 58

Experimental Neurology 299 (2018) 56–64

O. Einstein et al.

groups we used the mixed-factor ANOVA test. This test identified differences between the two independent groups, while subjecting mice to repeated “within-subjects” measures of their disease score. For performance tests, muscle enzyme activities, clinical and pathological parameters, the experimental groups were compared using two-tailed Student's t-test. For the in vitro experiments the experimental groups were compared using Student's paired t-test. Data were analyzed in IBM SPSS Statistics v.22. Differences were considered significant at p < 0.05.

perivascular infiltrates and parenchyma, and represented as total average number of each cell type per square millimeter. Demyelination was assessed by calculating the area of LFB loss, representing areas of myelin destruction. For the quantification of acute axonal damage, APP + axonal swellings and spheroids were counted, and the average of APP + profiles per square millimeter was calculated. In addition, the area of reduced axonal density in Bielschowsky silver staining was also assessed. All pathological measurements were performed by using the Image J software analysis (ver. 1.51H, National Institute of Health, USA).

3. Results 2.7. In vitro activation and proliferation assays of lymph node cells (LNCs) derived from trained or sedentary PLP immunized mice

3.1. Exercise training improves physical performance and increases citrate synthase (CS) activity in soleus muscle

Lymph nodes (LN) were excised from trained and sedentary mice at 10 days after PLP immunization. LNCs were cultured as single cell suspensions with 10 μg/ml PLP peptide or 2.5 μg/ml concanavalin A (ConA) or with no stimulation, as previously described (Einstein et al. 2007). The proliferation of T cells obtained from sedentary and trained, PLP immunized mice was evaluated by fluorescent activated cell sorter (FACS) analysis for bromodeoxyuridine (BrdU) incorporation and for the incorporation of the cell division tracking dye 5(6)-carboxyfluorescein diacetate succinimidyl ester (CFSE), as described previously (Einstein et al. 2007). For BrdU incorporation, ConA and PLP-activated LNCs were pulsed after 48 h in culture with 20 μM BrdU for 1 h, then collected and stained with phycoerythrin (PE)-conjugated anti-CD3 (eBiosciense) followed by staining with fluorescein isothiocyanate (FITC)–conjugated anti-BrdU (BD Pharmingen). Non-pulsed activated cells were used as control samples. The fraction of CD3 +, BrdU+ cells (relative to total) was calculated by using the Stimulation Index parameter: percent of BrdU+ cells following activation divided by percent of BrdU + cells with no activation. For CFSE FACS analysis, LNCs were pulsed with 5 μM CFSE (eBioscience) for 10 min., washed, and further cultured with or without ConA for 72 h. CFSE-labeled, non-activated cells were used as control samples. The mean number of cycles of T cells that entered cell cycle was calculated. T-cell activation was analyzed by FACS following staining with FITC-labeled anti-CD4 (eBiosciense) and APC-labeled anti-CD25 (eBiosciense). The fraction of CD4 +, CD25 + cells out of total was calculated. Staining with PE-labeled anti-FoxP3 (eBiosciense) was used to identify the fraction of regulatory CD4 +, CD25 +, FoxP3 + T cells by FACS analysis. In all FACS experiments, cells were pre-coated with anti–mouse CD16/CD32 (BD Pharmingen), as an Fc blocker, to block non-specific binding. In early experiments we tested our antibodies with an isotype control, indicating their specificity. All samples were analyzed in a Cytomics FC 500 apparatus (Beckman Coulter, Life Science) using the CXP analysis software (ver. 2.1; Informer Technologies, Inc).

To examine the efficacy of the training protocol, performance tests and enzyme activity analysis were performed. In the trained group there was a significant increase in maximal speed of 10% and a significant increase in exercise tolerance of 60%, compared to sedentary mice (Table 1). ET increased CS activity in soleus muscle by 30%, but did not affect CS activity in EDL muscle, nor did it affect HAD activity (Table 1). 3.2. Exercise training induces systemic immunomodulation in donor mice to attenuate the clinical course of transfer EAE, but does not induce direct neuro-protective effects in trained EAE recipient mice To study the effects of ET on the systemic autoimmune process we examined the clinical and pathological severity of EAE induced in recipient naïve SJL mice, following transfer of LN-T cells from trained(trained-transfer EAE) vs. sedentary (sedentary-transfer EAE) donor mice (Fig. 1A). The average day of disease onset was similar among the two experimental groups and occurred between days 7–10 post LN-T cell transfer (trained-transfer EAE = 7.9 ± 1.0, n = 16; sedentarytransfer EAE = 8.9 ± 0.5, n = 13; p > 0.05). However, while the transfer of LN-T cells derived from PLP immunized-sedentary mice induced a severe clinical course of EAE in all of the mice, the transfer of LN-T cells derived from PLP immunized-trained mice caused a markedly attenuated EAE in recipient mice (Fig. 2A). There was a significant difference in the overall clinical score between the two experimental groups over 50 days post-LN-T cell transfer. In addition, the maximal clinical score (MCS) of initial disease episode significantly decreased by approximately 30% and there was over 50% reduction in the burden of disease (BOD; area under curve; Fig. 2B). Next we examined whether treadmill running program of recipient mice affected the severity of EAE via direct neuro-protective effects on the CNS. EAE was induced in trained (trained EAE) and sedentary (sedentary EAE) recipients by transfer of encephalitogenic LN-T cells that were obtained from another group of PLP immunized mice (Fig. 1B). Transfer of encephalitogenic PLP reactive LN-T cells to sedentary mice induced a severe clinical course of EAE in all mice (Fig. 2C). Training did not significantly affect the clinical course of EAE (Fig. 2C), nor MCS or BOD (Fig. 2D).

2.8. Cytokine gene determination of PLP-reactive lymph node cells (LNCs) Total RNA was prepared using the RNeasy Plus Mini Kit (QIAGEN) from LNCs that were excised from trained and sedentary mice (n = 3–5/group) at 10 days after PLP immunization or following their activation in vitro with PLP peptide. Complementary DNA was prepared from 0.5 ng total RNA using qScript cDNA Synthesis Kit (Quanta Biosciences), according to the manufacturer's instructions. Semiquantitative real-time polymerase chain reaction (PCR) was performed using the PerfeCTa SYBR Green FastMix, ROX (Quanta Biosciences).

3.3. Proteolipid protein (PLP) reactive lymph node-T cells from trained mice induces milder inflammation and tissue pathology in recipient EAE mice We further examined the neuropathological consequences of ET in recipient mice at day 14 after transfer of PLP reactive LN-T cells obtained from sedentary (Fig. 3A, D, E, I, J) or trained mice (Fig. 3B, F, G, K, L). APP immunohistochemistry (Fig. 3A, B) and Bielschowsky staining (Fig. 3D–G) indicated a significant reduction in acute axonal injury in the spinal cord white matter of trained-transfer EAE vs. sedentary-transfer EAE mice. This was indicated by a reduction of approximately 80 and 50% in the number of APP+ axons and in the area of axonal loss, respectively (Fig. 3C, H). LFB staining (Fig. 3I-L) showed

2.9. Statistical analyses All data are presented as mean ± standard error of mean (SE) of four to sixteen mice/group and are representative of three to four independent experiments. To compare the clinical course along 50 days post LN-T cell transfer between trained and sedentary experimental 59

Experimental Neurology 299 (2018) 56–64

O. Einstein et al.

Fig. 2. Exercise training inhibits the encephalitogenicity of lymph node (LN) T cell-derived from proteolipid (PLP) - immuunized mice, but does not induce CNS protective mechanisms in transfer model of experimental autoimmune encephalomyelitis (EAE). Clinical course (A) and clinical parameters (B) of transfer EAE in mice that received PLP-reactive LN T cells from trained mice (trained- transfer EAE) or from sedentary mice (sedentarytransfer EAE). Clinical course (C) and clinical parameters (D) of transfer EAE in trained mice (trained EAE) and sedentary mice (sedentary EAE). Transfer of LN T cells derived from trained, PLP-immunized mice to naïve recipients induced a significantly milder EAE course (A, B). Exercise training did not affect the clinical course of EAE in trained mice that were injected with encephalitogenic PLP- reactive LN-T cells (C, D). The severity of EAE was scored according to a 0–6 scale. MCS – maximal clinical score; BOD – burden of disease (area under curve). Data are represented as mean ± SE. ⁎p < 0.05.

Fig. 3. Attenuation of pathological parameters of experimental autoimmune encephalomyelitis (EAE) mice injected with proteolipid (PLP)- reactive LN-T cells from trained mice. Pathological evaluation of axonal damage (A-H) and demyelination (I-M) was performed in the spinal cords of EAE mice that were injected with LN-T cells from control sedentary mice (A, D, E, I, J; n = 6) or with LN-T cells from trained mice (B, F, G, K, L; n = 6). D, F, I, K - dashed squares: represent areas of tissue damage shown in E, G, J, L, respectively. C, H, M – measurements of tissue pathology in spinal cord white matter of sedentary- transfer EAE (Sed-tr EAE) mice and trained- transfer EAE (Trained-tr EAE) mice. In Trained-tr EAE there were less amyloid precursor protein (APP) + injured axons (B, arrows) than in Sed-tr EAE (A, arrows; C). Bielschowsky staining showed less axonal damage and axonal loss in Trained-tr EAE mice (F, G) than in control Sed-tr EAE mice (D, E; H). Luxol fast blue staining showed reduction in the area of demyelination in Trained-tr EAE (K, L) vs. Sed-tr EAE EAE mice (I, J; M). Data are represented as mean ± SE. ⁎p < 0.05. Scale bars = 100 μm. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

60

Experimental Neurology 299 (2018) 56–64

O. Einstein et al.

Fig. 4. Attenuation of inflammatory parameters of experimental autoimmune encephalomyelitis (EAE) mice injected with proteolipid (PLP)- reactive LN-T cells from trained mice. Pathological evaluation of inflammation was performed in the spinal cords of EAE mice that were injected with LN-T cells from control sedentary mice (A, D, G; n = 6) or with LN-T cells from trained mice (B, E, H; n = 6), at the peak of the acute phase of EAE. C, F, I – counts of inflammatory cell types in spinal cords of sedentary- transfer EAE (Sed-tr EAE) and trainedtransfer EAE (Trained-tr EAE) mice. In Trained-tr EAE mice there was a significant reduction in total perivascular immune cell infiltrations (B), in CD3 + T cells (E) and in Mac3 + macrophages (H) vs. Sed-tr EAE mice (A, D, G, respectively). Data are represented as mean ± SE. ⁎p < 0.05. Scale bars: A, B, D, E, G, H = 100 μm.

from sedentary mice (Fig. 5B). However, there was no significant difference in the fraction of CD4 +, CD25 +, FoxP3 + regulatory T cells in LN excised at 10 days post PLP immunization between groups (Fig. 5B). There was 17–48-fold increase (in different experiments) in the expression of IL-10 in LNCs obtained from PLP immunized trained mice vs. sedentary controls. There were no significant differences in mRNA levels of other cytokines between groups. (Fig. 5C). We further characterized the inhibitory effect of ET specifically on T cells following 72 h of stimulation in vitro, and examined their encephalitogenic potential prior to their transfer to naïve recipients to induce EAE. ET significantly inhibited PLP stimulated T-cell proliferation in response to PLP, as indicated by approximately 50% reduction in the fraction of CD3 +, BrdU + T cells (Fig. 5E). Additionally, following in vitro stimulation with PLP of T cells derived from PLP immunizedtrained mice there was a 2.5- to 7-fold decrease in mRNA levels of all measured cytokines as compared to sedentary mice (Fig. 5D). To investigate whether the inhibitory effect of ET on LN-T cells was specific to the PLP auto-antigen or was a general inhibitory effect, we further performed FACS analysis for BrdU incorporation in T cells that were derived from PLP immunized-trained and sedentary mice and challenged in vitro with the non-specific mitogen ConA. Here we found that in response to ConA in vitro, ET not only did not inhibit the proliferation of PLP stimulated T-cells, but rather markedly increased proliferation (> 50%, Fig. 5F). Finally, we studied the effect of ET on T-cell division cycles. To this end, FACS analysis for CFSE content was performed on LN-T cells derived from PLP-immunized trained mice and sedentary mice following stimulation in vitro with ConA. In T cells (that entered cell cycle) obtained from PLP immunized trained mice, there was almost a 30% reduction in the mean number of cycles, compared to the number of T cells cycles in control PLP-immunized sedentary mice (Figs. 5G, H).

that in trained-transfer EAE mice there was a 40% decrease in the area of demyelination vs. control sedentary-transfer EAE mice (Fig. 3M). Since the severity of tissue damage in EAE is related to the destructive autoimmune inflammatory process (Linker et al. 2005; Martin and McFarland 1995) we tested whether transfer of LN-T cells from trained mice affected CNS inflammation in the recipient mice. Indeed, the attenuation in tissue pathology was supported by a markedly decreased inflammatory process in trained-transfer EAE vs. sedentarytransfer EAE control (Fig. 4). The attenuated inflammation in trainedtransfer EAE spinal cords (Figs. 4B, E, H) vs. control sedentary-transfer EAE cords (Figs. 4A, D, G) was indicated by a 50% overall decrease in total perivascular immune cell infiltrations (Fig. 4C), numbers of CD3 + T-cells (Fig. 4F) and number of Mac3 + macrophage (Fig. 4I) counts, infiltrating the spinal cord. 3.4. Exercise training inhibits the activation and proliferation of lymph node (LN)-T cells derived from PLP–immunized mice in vitro The in vivo experiments showed that ET reduces the potency of LN-T cells to induce brain inflammation. We therefore hypothesized that ET may actively interfere with the generation of effector T cells. Mice were subjected to the ET training and were immunized with PLP as described. LNCs were harvested from mice at 10 days post PLP immunization and cultured in vitro. LNCs derived from PLP immunized sedentary mice served as controls. We first measured the total number of LNCs obtained at 10 days after PLP immunization in trained vs. sedentary immunized mice. When the same number of LN was excised, ET significantly reduced the total number of LNCs by almost 50% (Fig. 5A). Furthermore, FACS analysis indicated that in the CD4 + T cell population excised from trained mice at 10 days after PLP immunization, there was a 20% reduction in the expression of the CD25 + activation marker in CD4 + cells vs. values 61

Experimental Neurology 299 (2018) 56–64

O. Einstein et al.

Fig. 5. Suppressive effects of exercise training on lymph node cells (LNCs) and T cells derived from proteolipid (PLP) immunized mice. LNCs were excised from trained and sedentary mice at 10 days post PLP immunization and were stimulated in vitro with proteolipid (PLP) peptide or ConcanavalinA (ConA). At day of LN excision the total number of LNCs (A) and the fraction of CD + C25+ cells (B) were significantly lower in trained mice. No significant difference was noted in the fraction of CD4 +, CD25 +, FoxP3 + cells (B) nor in the mRNA levels of tumor necrosis factor (TNF)- α, interferon (IFN)- γ, interleukin (IL)-17, IL-4 and transforming growth factor (TGF)-β between the groups, except for an increase in the IL-10 (C). After 72 h of stimulation in vitro, there was a decrease in all cytokines mRNA expression in the trained group (D). In addition, there was a significant reduction in Bromodeoxyuridine (BrdU) incorporation into the CD3+ T cells derived from trained mice in response to PLP peptide (E), but a significant elevation in response to ConA (F). Fluorescent-activated cell sorter (FACS) analysis at 72 h after 5(6)-carboxyfluorescein diacetate succinimidyl ester (CFSE) labeling indicated a significant reduction in the mean number of cycles of T cells derived from trained vs. sedentary mice, in response to ConA (G, H). Light gray – sedentary mice, dark gray – trained mice. A, B, E, F: Summary of three/ four independent experiments. Data are represented as mean ± SE. ⁎p < 0.05, ⁎⁎p < 0.01. C, D, H: Representatives of one of three independent experiments. C and D – relative expression to sedentary group = 1.

4. Discussion

immunization with the PLP peptide. Several lines of evidence in our study demonstrate that ET suppresses T-cell activation and proliferation in response to PLP, and inhibits their potency, but does not shift them to a regulatory cell population upon immunization with an auto-antigen. Immunomodulatory effects of ET were also documented in other experimental systems. It was shown that exercise training can reduce the levels of bacterial lipopolysaccharide (LPS)-stimulated secretion of pro-inflammatory cytokines (Stewart et al. 2005), reduce C-reactive protein (CRP) levels (Gleeson et al. 2011; McFarlin et al. 2006; Michigan et al. 2011) and increase circulating levels of the anti-inflammatory cytokine IL-1 receptor antagonist (IL-1ra) (Gleeson et al. 2011; McFarlin et al. 2006; Starkie et al. 2003; Steensberg 2003). Furthermore, exercise was shown to reduce the number of macrophages infiltrating adipose tissue in obese rodents (Kawanishi et al. 2013) and change the phenotype of resident adipose tissue macrophages from proinflammatory (M1) to anti-inflammatory (M2) (Kawanishi et al. 2010). Several mechanisms were suggested to be involved in the contribution of ET to an anti-inflammatory state in the periphery (Timmerman et al. 2008), such as the release of IL-6 from contracting skeletal muscle (Keller et al. 2003), and reduction in the expression of toll-like receptor (TLR) 4 (Flynn and McFarlin 2006; Gleeson et al. 2011; McFarlin et al. 2006; Robinson et al. 2015; Stewart et al. 2005). Physiological inhibition of T cells is considered to occur either by their deletion or by their suppression, wherein surviving T cells are unable to respond to antigenic stimuli (Van Parijs and Abbas 1998). On the other hand, it is well established that ET actually improves the ability of the immune system to respond to deleterious stimuli (Cao Dinh et al. 2017; Gleeson et al. 2011; Walsh et al. 2011). Therefore, an important issue in our experimental set-up is whether the LN-T cells of trained mice that exhibited inhibited responses to the PLP keep their ability to respond to other non-specific stimuli. In this context we found that in LNCs obtained from trained mice there was a significantly higher expression of IL-10 mRNA, compared to LNCs from sedentary controls. Moreover, we found a striking increase in the proliferative response of LN-T cell derived from trained PLP– immunized mice following stimulation in vitro with ConA. This demonstrates that while ET inhibits the immune responses to the myelin autoantigen, it does not induce general immune-suppression and does not decrease the maximal capacity of immune response to a non-specific mitogen. This important

The major findings of the present study are that exercise training: 1. attenuates EAE by modulating the systemic immune system; 2. increases immune responses to a non-specific stimulus; and 3. does not protect the CNS from encephalitogenic T cells. Earlier studies indicated beneficial effects of exercise on the clinical symptoms associated with EAE. This was correlated with preservation of axons and motor neurons, reduction in immune-cell infiltration, reduction in demyelination and axonal injury, increased synaptic plasticity, upregulation of neurotrophins and induction of hippocampal neurogenesis, inhibition of pro-inflammatory cytokine production, antioxidant effects and restoration of tight junction expression in the spinal cord (Benson et al. 2015; Bernardes et al. 2016; Bernardes et al. 2013; Kim and Sung 2017; Patel and White 2013; Pryor et al. 2015; Rossi et al. 2009; Souza et al. 2016). Others found no significant effects of exercise in relapsing remitting EAE (Klaren et al. 2016). However, one of the key issues is whether the suggested therapeutic effect in EAE is mediated directly in the CNS or systemically. Previous studies mostly used active EAE models, namely the MOG EAE in C57BL/6 mice (Benson et al. 2015; Bernardes et al. 2016; Bernardes et al. 2013; Klaren et al. 2016; Patel and White 2013; Pryor et al. 2015; Rossi et al. 2009; Souza et al. 2016), that could not distinguish between an effect of ET mediated via the systemic immune system to reduce encephalitogenicity vs. a direct protective effect on the CNS. We utilized here the PLP transfer EAE model in SJL mice: First, it enabled making this distinction and clearly showed a modulatory effect of ET on systemic autoimmunity rather than neuroprotection. Second, in this model injected LN-T cells from donor mice attack the CNS of recipient mice, leading to a relapsing remitting disease, thus mimicking the common clinical course of human MS. We show here that transfer of PLP reactive LN-T cells obtained from trained mice reduce the immune cell infiltrations, resulting in reduced demyelination and axonal pathology and improved clinical outcome in recipient mice. Both induction of immune responses and modulation of T-cell reactivity to brain-derived antigens take place in peripheral lymphoid organs (de Vos et al. 2002; Phillips et al. 1997; Widner et al. 1988). We therefore hypothesized that ET inhibits the generation of effector LN-T cells in the peripheral lymphoid system upon 62

Experimental Neurology 299 (2018) 56–64

O. Einstein et al.

finding is in accordance with a suggested model of “bio-regulatory effect of PE” (Ortega 2016), wherein ET reduces or prevents any excessive effects of selected inflammatory mediators in certain conditions, but still allows defenses in other conditions. Interestingly, while there was an increase in T cell proliferation obtained from trained mice in response to ConA, there was a reduction in the number of cycles of these cells. This may be due to a strong initial activation, followed by a reduced cycling capacity of the cells. It is well established that chronic stress results in immunosuppression and amelioration of symptoms associated with MS/EAE (Heesen et al. 2007). Acute activation of the sympathetic nervous system and the hypothalamic-pituitary-adrenal (HPA) axis is a necessary feature of ET-induced stress (Campisi et al. 2003; Girard and Garland, 2002). Stimulation of these systems may potentially affect some aspects of the immune response (Dhabhar 2002). Thus it could be argued that stress and immunosuppression associated with ET, rather than other adaptations, may mediate some of the positive clinical effects observed in our study. At present we cannot determine the extent to which these factors impacted on EAE progression. However, the finding that maximal T-cell proliferation (in response to ConA) was increased by ET speaks against the idea of general immunosuppression as mediator of beneficial effects of exercise on EAE progression. Additional studies are required to identify the mechanisms of ET on progression of EAE. The health benefits of ET may extend to the CNS as well. ET is associated with enhanced neurogenesis and neuroplasticity in various brain pathological conditions, as well as in increased expression of various neurotrophic factors (Baek 2016; Uysal et al. 2015; Voss et al. 2013). Recent studies suggested that exercise can modify CNS inflammation and specifically induce an anti-neuroinflammatory phenotype similar to its effects on the peripheral immune system (Kohman et al. 2012). However, our training protocol did not alter the clinical course of EAE, suggesting that no direct neuroprotective effects occurred in the CNS.

Phys. 284, R520–530. Cao Dinh, H., Beyer, I., Mets, T., Onyema, O.O., Njemini, R., Renmans, W., De Waele, M., Jochmans, K., Vander Meeren, S., Bautmans, I., 2017. Effects of physical exercise on markers of cellular immunosenescence: a systematic review. Calcif. Tissue Int. 100, 193–215. Dhabhar, F.S., 2002. Stress-induced augmentation of immune function—the role of stress hormones, leukocyte trafficking, and cytokines. Brain Behav. Immun. 16, 785–798. Einstein, O., Fainstein, N., Vaknin, I., Mizrachi-Kol, R., Reihartz, E., Grigoriadis, N., Lavon, I., Baniyash, M., Lassmann, H., Ben-Hur, T., 2007. Neural precursors attenuate autoimmune encephalomyelitis by peripheral immunosuppression. Ann. Neurol. 61, 209–218. Florindo, M., 2014. Inflammatory cytokines and physical activity in multiple sclerosis. ISRN Neurol 2014, 151572. Flynn, M.G., McFarlin, B.K., 2006. Toll-like receptor 4: link to the anti-inflammatory effects of exercise? Exerc. Sport Sci. Rev. 34, 176–181. Frohman, E.M., Racke, M.K., Raine, C.S., 2006. Multiple sclerosis–the plaque and its pathogenesis. N. Engl. J. Med. 354, 942–955. Girard, I., Garland Jr., T., 2002. Plasma corticosterone response to acute and chronic voluntary exercise in female house mice. J. Appl. Physiol. 1985 (92), 1553–1561. Gjevestad, G.O., Holven, K.B., Ulven, S.M., 2015. Effects of exercise on gene expression of inflammatory markers in human peripheral blood cells: a systematic review. Curr Cardiovasc Risk Rep 9, 34. Gleeson, M., Bishop, N.C., Stensel, D.J., Lindley, M.R., Mastana, S.S., Nimmo, M.A., 2011. The anti-inflammatory effects of exercise: mechanisms and implications for the prevention and treatment of disease. Nat. Rev. Immunol. 11, 607–615. Heesen, C., Gold, S.M., Huitinga, I., Reul, J.M., 2007. Stress and hypothalamic-pituitaryadrenal axis function in experimental autoimmune encephalomyelitis and multiple sclerosis - a review. Psychoneuroendocrinology 32, 604–618. Hemmer, B., Kerschensteiner, M., Korn, T., 2015. Role of the innate and adaptive immune responses in the course of multiple sclerosis. Lancet Neurol. 14, 406–419. Kawanishi, N., Yano, H., Yokogawa, Y., Suzuki, K., 2010. Exercise training inhibits inflammation in adipose tissue via both suppression of macrophage infiltration and acceleration of phenotypic switching from M1 to M2 macrophages in high-fat-dietinduced obese mice. Exerc. Immunol. Rev. 16, 105–118. Kawanishi, N., Mizokami, T., Yano, H., Suzuki, K., 2013. Exercise attenuates M1 macrophages and CD8+ T cells in the adipose tissue of obese mice. Med. Sci. Sports Exerc. 45, 1684–1693. Keller, P., Keller, C., Carey, A.L., Jauffred, S., Fischer, C.P., Steensberg, A., Pedersen, B.K., 2003. Interleukin-6 production by contracting human skeletal muscle: autocrine regulation by IL-6. Biochem. Biophys. Res. Commun. 310, 550–554. Kim, T.W., Sung, Y.H., 2017. Regular exercise promotes memory function and enhances hippocampal neuroplasticity in experimental autoimmune encephalomyelitis mice. Neuroscience 346, 173–181. Klaren, R.E., Stasula, U., Steelman, A.J., Hernandez, J., Pence, B.D., Woods, J.A., Motl, R.W., 2016. Effects of exercise in a relapsing-remitting model of experimental autoimmune encephalomyelitis. J. Neurosci. Res. 94, 907–914. Kohman, R.A., DeYoung, E.K., Bhattacharya, T.K., Peterson, L.N., Rhodes, J.S., 2012. Wheel running attenuates microglia proliferation and increases expression of a proneurogenic phenotype in the hippocampus of aged mice. Brain Behav. Immun. 26, 803–810. Kuerten, S., Angelov, D.N., 2008. Comparing the CNS morphology and immunobiology of different EAE models in C57BL/6 mice - a step towards understanding the complexity of multiple sclerosis. Ann. Anat. 190, 1–15. Linker, R.A., Sendtner, M., Gold, R., 2005. Mechanisms of axonal degeneration in EAE– lessons from CNTF and MHC I knockout mice. J. Neurol. Sci. 233, 167–172. Martin, R., McFarland, H.F., 1995. Immunological aspects of experimental allergic encephalomyelitis and multiple sclerosis. Crit. Rev. Clin. Lab. Sci. 32, 121–182. McFarland, H.F., Martin, R., 2007. Multiple sclerosis: a complicated picture of autoimmunity. Nat. Immunol. 8, 913–919. McFarlin, B.K., Flynn, M.G., Campbell, W.W., Craig, B.A., Robinson, J.P., Stewart, L.K., Timmerman, K.L., Coen, P.M., 2006. Physical activity status, but not age, influences inflammatory biomarkers and toll-like receptor 4. J. Gerontol. A Biol. Sci. Med. Sci. 61, 388–393. Michigan, A., Johnson, T.V., Master, V.A., 2011. Review of the relationship between Creactive protein and exercise. Mol. Diagn. Ther. 15, 265–275. Motl, R.W., Pilutti, L.A., 2012. The benefits of exercise training in multiple sclerosis. Nat. Rev. Neurol. 8, 487–497. Ortega, E., 2016. The “bioregulatory effect of exercise” on the innate/inflammatory responses. J. Physiol. Biochem. 72, 361–369. Patel, D.I., White, L.J., 2013. Effect of 10-day forced treadmill training on neurotrophic factors in experimental autoimmune encephalomyelitis. Appl. Physiol. Nutr. Metab. 38, 194–199. Phillips, M.J., Needham, M., Weller, R.O., 1997. Role of cervical lymph nodes in autoimmune encephalomyelitis in the Lewis rat. J. Pathol. 182, 457–464. Pilutti, L.A., Platta, M.E., Motl, R.W., Latimer-Cheung, A.E., 2014. The safety of exercise training in multiple sclerosis: a systematic review. J. Neurol. Sci. 343, 3–7. Pruimboom, L., Raison, C.L., Muskiet, F.A., 2015. Physical activity protects the human brain against metabolic stress induced by a postprandial and chronic inflammation. Behav. Neurol. 2015, 569869. Pryor, W.M., Freeman, K.G., Larson, R.D., Edwards, G.L., White, L.J., 2015. Chronic exercise confers neuroprotection in experimental autoimmune encephalomyelitis. J. Neurosci. Res. 93, 697–706. Qi, Z., He, J., Su, Y., He, Q., Liu, J., Yu, L., Al-Attas, O., Hussain, T., Ding, S., Ji, L., Qian, M., 2011. Physical exercise regulates p53 activity targeting SCO2 and increases mitochondrial COX biogenesis in cardiac muscle with age. PLoS One 6, e21140. Ritchie, I.R., MacDonald, T.L., Wright, D.C., Dyck, D.J., 2014. Adiponectin is sufficient,

5. Conclusions Exercise-mediated attenuation of CNS inflammation and clinical disease of EAE derives from a systemic immunomodulatory effect. These findings may have clinical relevance, where repeated CNS attack by systemic encephalitogenic lymphocytes underlies the pathogenic course of relapsing-remitting MS (Hemmer et al. 2015). Thus, ETmediated immunomodulation could prove useful in restraining the severity of MS attacks. Further studies are required to elucidate the cellular and molecular mechanisms behind ET-mediated immunomodulation. Acknowledgements This work was supported in part by the Judy and Sidney Swartz Foundation and by Ariel University. References Baek, S.S., 2016. Role of exercise on the brain. J Exerc Rehabil 12, 380–385. Benson, C., Paylor, J.W., Tenorio, G., Winship, I., Baker, G., Kerr, B.J., 2015. Voluntary wheel running delays disease onset and reduces pain hypersensitivity in early experimental autoimmune encephalomyelitis (EAE). Exp. Neurol. 271, 279–290. Bernardes, D., Oliveira-Lima, O.C., Silva, T.V., Faraco, C.C., Leite, H.R., Juliano, M.A., Santos, D.M., Bethea, J.R., Brambilla, R., Orian, J.M., Arantes, R.M., CarvalhoTavares, J., 2013. Differential brain and spinal cord cytokine and BDNF levels in experimental autoimmune encephalomyelitis are modulated by prior and regular exercise. J. Neuroimmunol. 264, 24–34. Bernardes, D., Brambilla, R., Bracchi-Ricard, V., Karmally, S., Dellarole, A., CarvalhoTavares, J., Bethea, J.R., 2016. Prior regular exercise improves clinical outcome and reduces demyelination and axonal injury in experimental autoimmune encephalomyelitis. J. Neurochem. 136 (Suppl. 1), 63–73. Campisi, J., Leem, T.H., Greenwood, B.N., Hansen, M.K., Moraska, A., Higgins, K., Smith, T.P., Fleshner, M., 2003. Habitual physical activity facilitates stress-induced HSP72 induction in brain, peripheral, and immune tissues. Am. J. Phys. Regul. Integr. Comp.

63

Experimental Neurology 299 (2018) 56–64

O. Einstein et al.

Behav. Immun. 19, 389–397. Timmerman, K.L., Flynn, M.G., Coen, P.M., Markofski, M.M., Pence, B.D., 2008. Exercise training-induced lowering of inflammatory (CD14 + CD16 +) monocytes: a role in the anti-inflammatory influence of exercise? J. Leukoc. Biol. 84, 1271–1278. Uysal, N., Kiray, M., Sisman, A.R., Camsari, U.M., Gencoglu, C., Baykara, B., Cetinkaya, C., Aksu, I., 2015. Effects of voluntary and involuntary exercise on cognitive functions, and VEGF and BDNF levels in adolescent rats. Biotech. Histochem. 90, 55–68. Van Parijs, L., Abbas, A.K., 1998. Homeostasis and self-tolerance in the immune system: turning lymphocytes off. Science 280, 243–248. de Vos, A.F., van Meurs, M., Brok, H.P., Boven, L.A., Hintzen, R.Q., van der Valk, P., Ravid, R., Rensing, S., Boon, L., t Hart, B.A., Laman, J.D., 2002. Transfer of central nervous system autoantigens and presentation in secondary lymphoid organs. J. Immunol. 169, 5415–5423. Voss, M.W., Vivar, C., Kramer, A.F., van Praag, H., 2013. Bridging animal and human models of exercise-induced brain plasticity. Trends Cogn. Sci. 17, 525–544. Walsh, N.P., Gleeson, M., Shephard, R.J., Gleeson, M., Woods, J.A., Bishop, N.C., Fleshner, M., Green, C., Pedersen, B.K., Hoffman-Goetz, L., Rogers, C.J., Northoff, H., Abbasi, A., Simon, P., 2011. Position statement. Part one: immune function and exercise. Exerc. Immunol. Rev. 17, 6–63. Wekerle, H., 2008. Lessons from multiple sclerosis: models, concepts, observations. Ann. Rheum. Dis. 67 (Suppl. 3), iii56–60. Widner, H., Moller, G., Johansson, B.B., 1988. Immune response in deep cervical lymph nodes and spleen in the mouse after antigen deposition in different intracerebral sites. Scand. J. Immunol. 28, 563–571. Zhang, S.J., Sandstrom, M.E., Lanner, J.T., Thorell, A., Westerblad, H., Katz, A., 2007. Activation of aconitase in mouse fast-twitch skeletal muscle during contractionmediated oxidative stress. Am. J. Phys. Cell Phys. 293, C1154–1159.

but not required, for exercise-induced increases in the expression of skeletal muscle mitochondrial enzymes. J. Physiol. 592, 2653–2665. Robinson, A.P., Harp, C.T., Noronha, A., Miller, S.D., 2014. The experimental autoimmune encephalomyelitis (EAE) model of MS: utility for understanding disease pathophysiology and treatment. Handb. Clin. Neurol. 122, 173–189. Robinson, E., Durrer, C., Simtchouk, S., Jung, M.E., Bourne, J.E., Voth, E., Little, J.P., 2015. Short-term high-intensity interval and moderate-intensity continuous training reduce leukocyte TLR4 in inactive adults at elevated risk of type 2 diabetes. J. Appl. Physiol. 1985 (119), 508–516. Rossi, S., Furlan, R., De Chiara, V., Musella, A., Lo Giudice, T., Mataluni, G., Cavasinni, F., Cantarella, C., Bernardi, G., Muzio, L., Martorana, A., Martino, G., Centonze, D., 2009. Exercise attenuates the clinical, synaptic and dendritic abnormalities of experimental autoimmune encephalomyelitis. Neurobiol. Dis. 36, 51–59. Souza, P.S., Goncalves, E.D., Pedroso, G.S., Farias, H.R., Junqueira, S.C., Marcon, R., Tuon, T., Cola, M., Silveira, P.C., Santos, A.R., Calixto, J.B., Souza, C.T., de Pinho, R.A., Dutra, R.C., 2016. Physical exercise attenuates experimental autoimmune encephalomyelitis by inhibiting peripheral immune response and blood-brain barrier disruption. Mol. Neurobiol. Spielman, L.J., Little, J.P., Klegeris, A., 2016. Physical activity and exercise attenuate neuroinflammation in neurological diseases. Brain Res. Bull. 125, 19–29. Starkie, R., Ostrowski, S.R., Jauffred, S., Febbraio, M., Pedersen, B.K., 2003. Exercise and IL-6 infusion inhibit endotoxin-induced TNF-alpha production in humans. FASEB J. 17, 884–886. Steensberg, A., 2003. The role of IL-6 in exercise-induced immune changes and metabolism. Exerc. Immunol. Rev. 9, 40–47. Stewart, L.K., Flynn, M.G., Campbell, W.W., Craig, B.A., Robinson, J.P., McFarlin, B.K., Timmerman, K.L., Coen, P.M., Felker, J., Talbert, E., 2005. Influence of exercise training and age on CD14 + cell-surface expression of toll-like receptor 2 and 4. Brain

64