In the present study, we evaluated stimulation of the angiotensin type 2 receptor (AT2R) by the selective non-peptide agonist Compound 21 (C21) as a novel therapeutic concept for the treatment of multiple sclerosis using the model of experimental autoimmune encephalomyelitis (EAE) in mice. C57BL-6 mice were immunized with myelin-oligodendrocyte peptide and treated for 4 weeks with C21 (0.3 mg/kg/day i.p.). Potential effects on myelination, microglia and T-cell composition were estimated by immunostaining and FACS analyses of lumbar spinal cords. The in vivo study was complemented by experiments in aggregating brain cell cultures and microglia in vitro.
In the EAE model, treatment with C21 ameliorated microglia activation and decreased the number of total T-cells and CD4+ T-cells in the spinal cord. Fluorescent myelin staining of spinal cords further revealed a significant reduction in EAE-induced demyelinated areas in lumbar spinal cord tissue after AT2R stimulation. C21-treated mice had a significantly better neurological score than vehicle-treated controls. In aggregating brain cell cultures challenged with lipopolysaccharide (LPS) plus interferon-γ (IFNγ), AT2R stimulation prevented demyelination, accelerated re-myelination and reduced the number of microglia. Cytokine synthesis and nitric oxide production by microglia in vitro were significantly reduced after C21 treatment. These results suggest that AT2R stimulation protects the myelin sheaths in autoimmune central nervous system inflammation by inhibiting the T-cell response and microglia activation. Our findings identify the AT2R as a potential new pharmacological target for demyelinating diseases such as multiple sclerosis.
Since the AT2R had been shown to act through anti-inflammatory and neuroprotective effects, we decided to test the effect of direct pharmacological stimulation of the AT2R in EAE in mice, which is a model of MS, i.e. a disease in which inflammation and neurodegeneration/demyelination are major components of the pathology.
Our study revealed that direct AT2R stimulation is anti-inflammatory in the CNS by attenuating T-cell recruitment and microglia activation. In addition, we found that AT2R stimulation protects from demyelination and promotes re-myelination–either directly or secondary to its anti-inflammatory, immune-modulatory effects. EAE-related neurological deficits were reduced.
Our findings identify AT2R stimulation as a potential new therapeutic concept for inflammatory, demyelinating diseases such as MS.
Multiple sclerosis (MS) is an autoimmune disease, in which the immune system attacks components of the myelin sheaths within the central nervous system (CNS) leading to demyelination, axonal loss and eventually functional impairment . It is the most common primary neurological disorder of young adults in most countries, affecting approximately 2.5 million people worldwide . Nowadays, treatment is able to delay disease progression and occurrence of relapses . As a result, life expectancy has improved significantly. However, disease progression can still only be prevented to a certain extent, and there is no cure. Drugs that are currently in use for the treatment of MS such as interferon (IFN)-β, glatiramer acetate and more recently fingolimod, laquinimod and natalizumab all aim at dampening down the immune response . The current search for new drugs is attempting to complement this armamentarium with better-tolerated immunomodulatory and with neuroprotective drugs .
The autoimmune response in MS is driven by myelin-specific Th1 and Th17 T-cells . Contact of these T-cells with local antigen-presenting cells in the CNS leads to T-cell reactivation and secretion of pro-inflammatory cytokines and chemokines, which in turn facilitate infiltration of further immune cells into the CNS. The antigen-presenting cells, which in the CNS mainly consist of microglia and to a lesser extent of infiltrating macrophages, are capable of releasing pro-inflammatory cytokines themselves, too . Once the inflammatory cascade is activated, it initiates the neurodegenerative response including destruction of oligodendrocytes and a demyelination process .
Several studies have provided evidence that the renin–angiotensin system (RAS) in the brain is involved in the pathogenesis of MS. An alteration in the levels of angiotensin II (AngII) and other components of the brain RAS–in particular angiotensinogen–has been demonstrated in cerebrospinal fluid from patients with MS [6–9]. Stegbauer et al.  reported an increased expression of renin, angiotensin-converting enzyme (ACE) and the angiotensin type 1 receptor (AT1R) in spinal cord tissue, spleen and leucocytes from animals suffering from myelin oligodendrocyte glycoprotein (MOG)-induced experimental autoimmune encephalomyelitis (EAE), an animal model closely resembling MS in humans. Lanz et al.  showed that AngII acting via the AT1R is essentially involved in CNS inflammation in the MOG-EAE model. Furthermore, treatment with the renin inhibitor aliskiren, the ACE inhibitor enalapril or the AT1R antagonist losartan resulted in a significant improvement in neurological outcome in diseased animals [10,12]. All of these therapeutic interventions aim at inhibiting actions of AngII mediated by the AT1R. Thus far, there have been no studies evaluating whether the angiotensin type 2 receptor (AT2R), which generally mediates actions opposing those of the AT1R [13,14], but also RAS-independent actions on growth factors or pro-inflammatory cytokines, may be of any therapeutic use in MS.
Regarding AT2R physiology in the CNS, it is generally accepted that the AT2R has neuroprotective properties resulting in the promotion of neurite outgrowth, in neuronal regeneration, improved cognitive function and a reduction in neurological deficits in models of stroke and–as we recently demonstrated–spinal cord injury [15–24]. Molecular mechanisms underlying AT2R-mediated neuroprotection appear to include anti-inflammatory and antioxidative actions as well as an induction of synthesis of the neurotrophin brain-derived neurotrophic factor (BDNF) and of methyl methanesulfonate sensitive 2 [23,25–28].
In the present study, in mice, we investigated whether pharmacological AT2R stimulation with a non-peptide, small molecule AT2R agonist, Compound 21 (C21), is able to ameliorate EAE pathology. We were particularly interested in the effect of AT2R stimulation on demyelination and on T-cell and microglia activation.
MATERIALS AND METHODS
Six-week-old female C57BL/6 mice (Janvier) were kept in a specific pathogen-free barrier under standardized conditions with respect to temperature and humidity and were housed on a 12-h light/12-h dark cycle in groups of five with food and water ad libitum. An acclimation period of at least 3 weeks was required prior to any experimental manipulation to get mice used to being handled and to reduce stress. Animal housing, care and applications of experimental procedures complied with the Guide for the Care and Use of Laboratory Animals of the State Government of Berlin, Germany and were approved by the State Government of Berlin.
Induction of EAE
On day 0 of the study, 9-week-old mice were immunized into four sites of their back (left and right shoulder, left and right flank) by subcutaneous (s.c.) injections with 250 μg of a peptide corresponding to amino acid residues 35–55 of mouse MOG (MOG35–55: MEVGWYRSPFSRVVHLYRNGK; Pepceuticals) in complete Freund's adjuvant (CFA; Difco) containing 0.8 mg of inactivated mycobacteria tuberculosis H37 RA (Difco). For the emulsion, MOG35–55 peptide and mycobacteria tuberculosis H37 RA were dissolved in PBS and mixed at a 1:1 ratio with CFA by the syringe-extrusion method until a stable emulsion was formed. On the day of immunization and 2 days later, 200 ng of pertussis toxin (Calbiochem) in 200 μl of PBS was applied intraperitoneally (i.p.). Sham mice receiving the immunization cocktail (Freund's adjuvant + mycobacteria tuberculosis) without MOG peptide served to control for effects of the immunization cocktail independent of the antigen.
The following numbers of animals were used: neurological evaluation, n=15 animals per group; FACS analysis, sham: n=4, EAE/vehicle: n=6, EAE/C21: n=9; immunohistochemistry of spinal cords, sham: n=2, EAE/vehicle: n=7, EAE/C21: n=7.
Clinical evaluation of EAE
Mice (n=15 per treatment group) were examined daily for clinical signs of EAE by two blinded observers and were scored as follows: 0, no clinical disease; 1, tail weakness or paralysis; 2, hindlimb weakness or paralysis; 3, complete hindlimb paralysis and forelimb weakness; 4, hindlimb paralysis and forelimb paralysis; 5, moribund or dead .
Body weight and overall condition of mice were monitored and evaluated daily. The following measures and humane endpoints were defined in order to limit suffering of animals: ≥5% loss in body weight or neurological score ≥1.5, supply of wet food; 5%−20% loss in body weight or neurological score ≥3, 0.9% NaCl+5% glucose s.c. twice per day in addition to wet food; >20% loss in body weight or neurological score >3.5, anaesthesia of mice with ketamine/xylazine (400/60 mg per kg i.p.; Sigma) followed by cervical dislocation. In the present study, one mouse of the EAE/vehicle group fulfilled the latter criteria and had to be killed on day 22 after immunization.
Animals were randomly assigned to treatment groups using the randomization function of Microsoft Excel. Investigators at any stage of the study/tissue analysis were blinded with regard to treatment group. The AT2R agonist C21 (0.3 mg/kg, kindly provided by Vicore Pharma) was applied i.p. once daily from day 3 before immunization until day 27 post-immunization (p.i.). The dose of C21 was chosen based on dose–response evaluations in a previous in vivo study . A C21 dose of 0.3 mg/kg i.p. in mice results in a maximal plasma peak concentration of approximately 0.5 μM. An equivalent volume of vehicle (aqua ad iniectabilia i.p. once daily; approximately 100 μl) was applied to control animals.
Tissue preparation for histology
On day 27 p.i., which corresponds to the chronic phase of the disease, mice were anaesthetized with ketamine/xylazine (400/60 mg/kg i.p.; Sigma) and intracardially perfused with 0.1 M PBS followed by 4% paraformaldehyde (PFA) in 0.1 M PBS. Spinal cords were removed, cryoprotected in 30% sucrose and serial cryostat longitudinal sections were cut at a 7 μm thickness.
Histological and immunohistochemical analyses
Acetone-fixed lumbar spinal cord cryosections were mounted in ProLong Gold antifade reagent (Invitrogen) and stained with FluoroMyelin™ Green Fluorescent Myelin Stain (Invitrogen) to evaluate myelin loss. The protocol was carried out as per the manufacturer's instructions. For visualization of cell nuclei, sections were washed with PBS and exposed to Hoechst 33342 (1:1000 dilution in PBS; Sigma) during 10 min at room temperature. Image J software was used for quantification. Three to five sections per animal were used for evaluation. For myelin quantification, images were converted into 8-bit grey-scale. The threshold range was set up between 10 and 20, depending on the intensity of each image and then converted to a binary format. A total section area was hand-traced measuring area and integrated density. To calculate the degree of myelination, the area containing FluoroMyelin™-labelled cells was calculated as a percentage of the total area that had been selected. Semi-quantitative histological evaluation was carried out for positively stained cell nuclei based on published protocols: score 0, no cell nuclei; score 1, cell nuclei present only in perivascular areas; score 2, low number of cell nuclei in the parenchyma; score 3, moderate number of cell nuclei in the parenchyma; and score 4, high number of cell nuclei in the parenchyma. All analyses were performed in a blinded manner.
Leucocyte isolation from the spinal cord, lymph nodes and spleen
Leucocytes were isolated from the lumbar part of the spinal cord, from lymph nodes and spleen on day 18 p.i., which is the time point of highest disease activity, i.e. most severe neurological deficits. Mice (n =4–9 animals/group) were anaesthetized with ketamine/xylazine (400/60 mg/kg i.p.; Sigma) and perfused intracardially with PBS. The lumbar part of the spinal cord (30 mm) was removed using hydraulic pressure from a 10 ml syringe with PBS, weighed and put on ice in RPMI 1640 (Gibco, Invitrogen). Lymph nodes and spleen were surgically excised. Samples were normalized according to size and weight. Tissues were surgically dissected, digested with collagenase (1 mg/ml; Sigma), mechanically homogenized and passed through a cell strainer (70 μm) to obtain a single-cell suspension. Homogenates were resuspended in 35% Easycoll (Biochrom) and underlaid with 70% Easycoll. The gradients were centrifuged at 600 g for 20 min without brake. Leucocytes were collected from the interface between the two Easycoll layers, passed through a cell strainer (70 μm), washed and resuspended in PBS containing 2% FBS (PAN Biotech).
Flow cytometric analysis
Mononuclear cells isolated from the lumbar part of the spinal cord, from lymph nodes and spleen on day 18 p.i. were incubated for 5 min at 4°C with mouse BD Fc BlockG™ (1:100 dilution) and stained for 20 min at 4°C with anti-CD8, anti-CD45, anti-CD11b, anti-CD4 and anti-CD25 (1:200 dilution; BD Pharmingen™; BD Biosciences). All antibodies were conjugated with different fluorochromes except anti-T-cell receptor (TCR) (1:100 dilution), for which a secondary antibody, streptavidin PerCP (1:200 dilution), was required (BD Pharmingen™; BD Biosciences). TCR and CD4 were used as markers for total T-cells or CD4+ T-cells, respectively. CD11b and CD45 were used as markers for detecting CD11b+CD45high cells, which include activated microglial cells and infiltrating macrophages . Resting microglia were identified as CD11b+CD45lowcells. Flow cytometric analysis was performed using a FACSCanto II Flow cytometer (Becton Dickinson) and data were analysed by FlowJo software (Tree Star).
Aggregating rat brain cell cultures
Aggregating brain cell cultures were prepared from the telencephalon of 16-day embryonic Sprague–Dawley rats (Hsd:SD, Harlan) as previously described in detail [32,33]. Briefly, the embryonic brain tissue was mechanically dissociated into single cells using cell strainers of 200 and 100 μm pores. Dissociated cells were incubated in chemically defined medium [Dulbecco's modified Eagle's medium (DMEM) adjusted for serum-free conditions as previously described in detail by Honneger et al. [31a] under constant gyratory agitation (80 rpm) at 37°C in an atmosphere of 10% CO2 and 90% humidified air. Medium was exchanged (replacement of 5 ml of a total of 8 ml) every third day from day 5 to day 14 and every other day in the following. To get identical samples of aggregates for experimentation, aggregates were pooled and redistributed into the required number of culture flasks prior to the experiment. Experiments were carried out starting on day 22 after cell isolation, when the myelination of axons was nearly maximal.
In vitro inflammation protocol
The inflammatory response and accompanying demyelination was triggered by the combined treatment with lipopolysaccharide (LPS; 5 μg/ml final concentration; Sigma) and IFNγ (50 units/ml final concentration; PeproTech) as described earlier . The treatment, initiated at day 22 in culture, was repeated twice, at day 24 and day 26. Stock solutions were prepared, for IFNγ in PBS containing 0.1% BSA (pH 8), and for LPS in 0.9% NaCl.
Prevention of demyelination
To test whether AT2R stimulation is able to prevent LPS/IFNγ-induced demyelination, aggregates were treated with LPS/IFNγ (5 μg/ml; 50 units/ml) on days 22, 24 and 26 after initiation of cultures and co-incubated daily with the AT2R agonist C21 (1 μM) or vehicle (same volume of culture medium) from day 22 to day 27. The AT2R antagonist PD123319 (10 μM; Sigma) was added daily simultaneously to C21 treatment. Cultures were harvested 24 h after the last treatment with C21 or vehicle.
Promotion of re-myelination
To test whether AT2R stimulation is able to promote re-myelination after LPS/IFNγ-induced demyelination, aggregates were first treated with LPS/IFNγ (5 μg; ml/50 units/ml) on days 22, 24 and 26 to induce demyelination. On day 28, aggregates were washed twice in culture medium and further maintained in culture for 7 days. Treatment with the AT2R agonist C21 (1 μM) or vehicle was performed daily from day 28 until day 34. The AT2R antagonist PD123319 (10 μM final concentration; Sigma) was added daily simultaneously to C21 treatment. Cultures were harvested 24 h after the last treatment with C21 or vehicle.
Histological analysis and immunocytochemistry of brain cell cultures
Aggregating rat brain cell cultures were harvested 1 day after the last treatment with C21 or vehicle. Briefly, they were washed twice with pre-warmed PBS, embedded in cryomatrix (Jung), frozen in isopentane, cooled with liquid nitrogen and stored at −80°C. For immunocytochemistry, cryosections (10 μm) were fixed for 20 min in 4% PFA dissolved in PBS at room temperature, and then washed in PBS. For blockade of non-specific binding, sections were first incubated in normal horse serum (1:25 dilution in PBS with 0.1% Triton X-100; Jackson Immuno Research Laboratories). They were then exposed overnight at 4°C to antibodies against myelin basic protein (MBP; mouse monoclonal, 1:40 dilution, Chemicon, Millipore) or MOG (mouse monoclonal,1:50 dilution, Chemicon, Millipore). Subsequently, sections were incubated with biotinylated horse anti-mouse IgG (1:200 dilution; Vector), followed by FITC-avidine (1:100 dilution; Vector). For the staining of microglia by the specific binding of FITC-conjugated isolectin B4 of Griffonia simplicifolia (IB-4), cryosections were washed for 15 min in Tris buffer containing 1% Triton X-100 and then incubated for 30 min in Image-IT FX signal enhancer (Invitrogen). Sections were then exposed overnight at 4°C to IB-4 (1:500 dilution; Sigma). Sections were mounted in ProLong Gold antifade reagent (Invitrogen). Image J software was used to quantify the positive labelled area of aggregate sections and at least 40 aggregate sections were analysed per condition. For myelin quantification, images were converted into 8-bit grey-scale. The threshold range was set up between 10 and 20, depending on the intensity of each image and then converted to a binary format. A total aggregate area was hand-traced measuring area and integrated density. To calculate the degree of myelination, the area containing MBP- or MOG-labelled cells was calculated as a percentage of the total area that had been selected.
Isolation, cultivation and stimulation of primary murine microglial cells
Primary microglial cells were prepared from newborn Sprague–Dawley rat pups. Meninges and choroid plexus membranes were removed from brains, and a hypothalamic block was dissected and minced with small scissors. The dissected tissues were further dissociated by filtering through a 100 μm pore nylon mesh (BD), followed by centrifuging (300 g; 5 min) at room temperature. Pellets were re-suspended in DMEM containing 10% FBS. The microglial cells were seeded on to poly-D-lysine-coated 100 mm dishes at a density of 1.5 × 105 cells/ml, and incubated with 95% air/5% CO2 at 37°C for 7 days without changing the medium. After this time period, floating microglia in the culture medium were collected and transferred to six-well plates for 3 h, and the original dishes were fed with fresh medium. Microglia were collected in this way at weekly intervals. Isolated cells were >95% pure microglia based on the co-localization of the microglia marker Iba1 with DAPI fluorescence. After attachment in the new plates, microglia were fed with fresh DMEM/10% FBS and incubated for another 2–3 days before each treatment. Cells were incubated for 24 h with C21 (0.1 μM) or vehicle (aqua ad injectabilia) followed by RNA isolation for quantitative reverse transcription (RT)-PCR analysis of IL-1β and Cd11b.
Cultivation and stimulation of the murine microglial cell line C8-B4
The murine microglial cell line C8-B4 (A.T.C.C.) was maintained at 37°C in 5% CO2 in DMEM (Invitrogen) with 4.5 g/l glucose, supplemented with 10% FBS and 1% penicillin/streptomycin (100 units/ml penicillin and 100 μg/ml streptomycin; PAN Biotech). Cells were split once a week and tested frequently for mycoplasma contamination. Cells were used up to passage 8.
C8-B4 microglial cells were cultured in serum-free DMEM in 25 cm2 flasks. Cells were incubated for 24 h with C21 (1 μM) or vehicle (aqua ad injectabilia). Afterwards, cells were incubated for an additional 6 h with C21 or vehicle in the presence of LPS (10 μg/ml) in order to analyse putative changes in mRNA expression of interleukin (IL)-6 and tumour necrosis factor (TNF)-α by quantitative RT-PCR.
mRNA isolation and cDNA synthesis
Standard RNA isolation from microglial cells was performed using Rneasy Kit (Qiagen), according to the manufacturer's protocol. RNA concentration was measured spectrophotometrically using Nanodrop® (PeqLab Biotechnologie) at a wavelength of 260 nm. Purity was assessed by the quotient 260/280 nm. Approximately 500 ng of total RNA was reverse transcribed using the High Capacity cDNA Reverse Transcription Kit (Applied Biosystems) in 20 μl reactions following the manufacturer's protocol. Reactions were carried out at 25°C for 10 min, 37°C for 120 min and 85°C for 5 min.
Real-time quantification for AT2R, IL-1β, Cd11b, IL-6 or TNF-α was performed using Taqman® Gene Expression Master Mix (Applied Biosystems) in accordance with the manufacturer's instructions. Real-time PCR reaction was run in an Mx3000p Real-Time PCR System (Stratagene/Agilent Technologies). The following Taqman® Gene Expression Primers were used: AT2R: Mm00431727_g1, Rn00560677_s1, IL-1β: Rn00580432_m1, Cd11b: Rn00709342_m1, IL-6: Mm00446190_m1, TNF-α: Mm00443260_g1, glyceraldehyde-3-phosphate dehydrogenase (GAPDH): Rn99999916_s1, β-actin: Mm00607939 (all from Applied Biosystems). The Taqman probe was labelled with FAM. Each quantitative RT-PCR included two no template controls per primer and one no reverse transcription control per sample to exclude genomic contamination. Data were analysed in triplicate in three independent runs (means±S.E.M.) via ΔΔCt analysis.
Nitric oxide production
C8-B4 microglial cells were washed with PBS, counted using Trypan Blue, resuspended in serum-free medium and placed in a 48-well plate at 3 × 105 cells in a volume of 300 μl. Cells were incubated for 24 h with C21 (1 μM) or vehicle (aqua ad injectabilia). Afterwards, cells were incubated for an additional 24 h with C21 or vehicle in the presence of LPS (10 μg/ml) plus IFNγ (100 units/ml). The AT2R antagonist PD123319 (10 μM) was added 30 min prior to C21. Supernatants were collected and assessed for nitric oxide (NO) production using the Griess Assay (Promega) following the manufacturer's protocol.
Results are expressed as means±S.E.M. or medians±range (semi-quantitative analyses). A Gaussian distribution was tested using the Kolmogorov–Smirnov test. Normal distributed data were analysed with one-way ANOVA followed by Bonferroni post hoc test. Non-parametric data were analysed using the Kruskal–Wallis test with Dunn's multiple comparison test. The analysis of the treatment effect on the neurological score (non-parametric scale) was carried out by Mann–Whitney U test. The assumption of equal variance was tested using Bartlett's test. Data sets with unequal variance were transformed to approximate the distribution to a Gaussian distribution. Differences were considered significant at P value <0.05. Statistical analyses were performed using GraphPad Prism 5 software for Windows.
AT2R stimulation attenuates demyelination and accumulation of cells in MOG-EAE diseased mice
Spinal cords were collected 27 days after immunization of mice with MOG peptide (chronic phase of the disease). Longitudinal sections of lumbar spinal cords were stained with FluoroMyelin™ for visualization of myelin and with Hoechst 33342 for visualization of cell nuclei. Successful induction of MOG-EAE caused significant demyelination when compared with sham mice (Figures 1A and 1B; P<0.01 vs. SHAM) and a pronounced increase in cell nuclei accumulation (Figures 1A and 1C). A co-localization of demyelinated areas and clusters of cells (cell nuclei) could be noted. MOG-EAE-diseased mice treated with the AT2R agonist C21 (0.3 mg/kg i.p.) showed a significant reduction in demyelinated areas (Figures 1A and 1B; P<0.01 vs. EAE/vehicle) and fewer clusters of cells as compared with the vehicle-treated MOG-EAE-diseased mice (Figures 1A and 1C); however, the latter did not reach statistical significance.
Histopathological examination of lumbar spinal cords
AT2R stimulation reduces T-cell infiltration in the spinal cord
In order to evaluate the nature of the cell clusters observed in EAE, mononuclear cells isolated from the lumbar part of the spinal cord on day 18 after immunization (inflammatory peak of the disease) were analysed using flow cytometry. We first gated cells for CD45high lymphocytes and then excluded CD11b+ cells from the monocytic lineage. TCR and CD4 were used as markers for T-cells and CD4+ T-cells (Figure 2A). Results revealed an increase in the absolute number of total T-cells (Figure 2B) and of CD4+ T-cells (Figure 2C) in EAE-diseased compared with healthy mice and a significant reduction in total T-cell and CD4+ T-cell number in C21-treated (0.3 mg/kg, i.p.) EAE-diseased mice when compared with vehicle-treated EAE mice (P<0.01 vs. EAE/vehicle).
Flow cytometric analysis of T-cells in lumbar spinal cord
There was no difference in the number of T-cells in lymph nodes and spleen from C21-treated compared with vehicle-treated EAE-diseased mice (data not shown).
AT2R stimulation reduces the number of resident and activated microglia in the spinal cord
Microglia and infiltrating macrophages are other types of pro-inflammatory cells in the CNS, which contribute to EAE pathology. We therefore investigated the effect of direct AT2R stimulation on resting microglia, activated microglia/CNS-infiltrating macrophages in lumbar spinal cord sections on day 18 p.i. using flow cytometry. The number of resting microglia (CD11b+CD45lowcells, Figures 3A and 3B) as well as activated microglia/CNS-infiltrating macrophages (CD11b+CD45high cells, Figures 3A and 3C) was significantly increased in MOG-EAE-diseased mice when compared with sham-immunised mice (P<0.01 vs. SHAM), but significantly reduced in animals treated with the AT2R agonist C21 (0.3 mg/kg, i.p.) when compared with vehicle-treated diseased mice (P<0.01, P<0.05 vs. EAE/vehicle).
Flow cytometric analysis of microglia/macrophages in lumbar spinal cord
AT2R stimulation attenuates neurological deficits in MOG-EAE-diseased mice
Clinical signs of autoimmune encephalomyelitis were scored and evaluated daily until day 27 p.i. by two independent blinded observers. Immunized mice–with the exception of two C21-treated mice–developed severe MOG-EAE with onset of paralysis occurring on day 12 p.i. (Figure 4 and Table 1). Symptoms peaked at day 18 and were followed by a stable chronic phase of disease.
|Group .||N .||Incidence (%=n/N) .||Disease onset .||Clinical score at disease peak, day 18 p.i. .||Clinical score at progressive stage, day 27 p.i. .|
|EAE/vehicle||15||100% (15/15)||day 12 p.i.||3 (0–3.5)||2.5 (0.75–5)|
|EAE/C21||15||86.66% (13/15)||day 12 p.i.||2.5 (0–3.5)||1.5 (0–3)*|
|Group .||N .||Incidence (%=n/N) .||Disease onset .||Clinical score at disease peak, day 18 p.i. .||Clinical score at progressive stage, day 27 p.i. .|
|EAE/vehicle||15||100% (15/15)||day 12 p.i.||3 (0–3.5)||2.5 (0.75–5)|
|EAE/C21||15||86.66% (13/15)||day 12 p.i.||2.5 (0–3.5)||1.5 (0–3)*|
Treatment with the AT2R agonist C21 (0.3 mg/kg, i.p.) starting 3 days before immunization significantly ameliorated the course of MOG-EAE in comparison with the vehicle-treated group (Figure 4; P<0.001 vs. EAE/vehicle). The positive effect of treatment was significant at the end of the study, i.e. on day 27 (Table 1), and when analysed over the whole course of the study. It already tended to be effective at day 18; however, this early effect was not yet statistically significant (Table 1).
AT2R stimulation protects against demyelination
Since demyelinated areas and cell clusters (presumably consisting of T-cells) co-localized in vivo pointing to demyelination being secondary to the immune response, we performed in vitro experiments in aggregating brain cell cultures, which are devoid of T-cells, in order to clarify whether AT2R stimulation is able to protect from demyelination in the absence of T-cells.
Treatment of aggregating rat brain cell cultures with LPS (5 μg/ml)/IFNγ (50 units/ml) over a period of 6 days resulted in significant demyelination as determined by immunohistochemical staining for both MBP and MOG (Figure 5; P<0.001 vs. vehicle-treated cultures). Co-incubation with C21 during the treatment with LPS/IFNγ significantly attenuated demyelination and maintained the degree of myelination at levels of unchallenged controls (Figure 5; P<0.001 vs. LPS/IFNγ-treated cultures). Co-incubation with the selective AT2R antagonist PD123319 significantly inhibited the protective effect of C21 on myelination suggesting an AT2R-specific effect of C21 (Figure 5; P<0.01, P<0.001 vs. LPS/IFNγ+C21-treated cultures). C21 given alone or in combination with PD123319 had no effect on the degree of myelination (data not shown).
Immunohistochemical staining of aggregating brain cell cultures: prevention of demyelination
AT2R stimulation promotes re-myelination
To investigate the effect of AT2R stimulation on re-myelination, demyelination was induced in aggregating rat brain cell cultures by LPS (5 μg/ml)/IFNγ (50 units/ml) treatment for 6 days. After replacement of medium and washing of cells, cultures were treated for an additional 7 days with C21 (1 μM) or vehicle in the absence of LPS/IFNγ. In vehicle-treated cells, demyelination was still persistent 7 days after termination of LPS/IFNγ (Figure 6; P<0.001 vs. controls not challenged with LPS/IFNγ). In contrast, when the AT2R agonist C21 was applied after termination of LPS/IFNγ, brain aggregates showed a significant degree of re-myelination and the degree of myelination was almost back to levels of un-challenged controls (Figure 6A: P<0.001; Figure 6B: P<0.05 vs. LPS/IFNγ-treated cultures). These effects of C21 were inhibited by the AT2R antagonist PD123319 pointing to AT2R-specificity of the C21 effects (Figure 6A: P<0.001; Figure 6B: P<0.05 vs. LPS/IFNγ + C21-treated cultures).
Immunohistochemical staining of aggregating rat brain cell cultures: promotion of re-myelination
AT2R stimulation decreases the number of microglia
Microglial cells were identified in aggregating brain cell cultures by immunocytochemical staining for IB-4. Treatment of aggregating rat brain cell cultures with LPS (5 μg/ml)/IFNγ (50 unib/ml) over a period of 6 days resulted in a significant increase in the number of microglial cells indicating microglial activation (Figure 7; P<0.001 vs. controls not challenged with LPS/IFNγ). Co-treatment with the AT2R agonist C21 in addition to LPS/IFNγ led to a significantly lower number of IB-4-positive cells when compared with vehicle-treated controls (Figure 7; P<0.001 vs. LPS/IFNγ-treated cultures). The AT2R antagonist PD123319 significantly blunted the effect of C21 (Figure 7; P<0.001 vs. LPS/IFNγ+C21-treated cultures).
Immunohistochemical staining of microglia within aggregating brain cell cultures
AT2R stimulation attenuates cytokine and NO release from microglia
NO and pro-inflammatory cytokines, such as IL-6, TNF-α and IL-1β, are important contributors to oligodendrocyte damage and consequently demyelination [5,35]. We therefore examined whether AT2R stimulation of microglial cells in vitro is able to inhibit expression/release of these mediators under basal or pro-inflammatory conditions.
In a first set of experiments, primary microglia isolated from the hypothalamus of Sprague–Dawley rats were treated with C21 (100 nM) or vehicle for 24 h under basal conditions. C21 treatment caused a significant reduction in CD11b mRNA expression, a marker for activated microglia. Expression of mRNA of the pro-inflammatory cytokine IL-1β was also significantly decreased in these cells (Figures 8A and 8B; P<0.001 vs. untreated group).
mRNA expression of pro-inflammatory cytokines in microglia
In a second set of experiments using the mouse microglial cell line C8-B4, we demonstrated that LPS (10 μg/ml)-induced expression of IL-6 was significantly reduced by AT2R stimulation with 1 μM C21 (Figure 8C; P<0.01 vs. LPS-treated cultures), whereas there was only a weak, non-significant reduction in TNF-α expression (Figure 8D).
NO is released by microglia and is able to affect oligodendrocytes . A 24 h incubation of C8-B4 microglia with the AT2R agonist C21 (1 μM) decreased LPS/IFNγ-induced NO release significantly when compared with vehicle-treated cells (Figure 9; P < 0.001 vs. LPS/IFNγ-treated cultures). This effect could be inhibited by blockade of the AT2R with PD123319 (10 μM; Figure 9; P<0.001 vs. LPS/IFNγ+C21-treated cultures).
NO release from microglia
AT2R expression in cells and tissues used in the present study
We confirmed by RT-PCR that microglia, T-cells, brain aggregates and spinal cord tissue express AT2R (Figure 10).
AT2R expression in cells and tissues used in the present study
The data presented herein constitute the first study in which the therapeutic effect of direct AT2R stimulation was evaluated in MOG-EAE, an animal model of MS. The major findings of the present study are that AT2R stimulation attenuates demyelination, T-cell infiltration and microglia activation in vivo coinciding with the amelioration of neurological deficits. Complementary in vitro studies in rat brain aggregating cell cultures confirmed the protection from demyelination and microglia activation and additionally revealed that AT2R stimulation accelerates re-myelination. Furthermore, we demonstrated an AT2R-mediated reduction in cytokine and NO release from isolated microglia in vitro.
Demyelination is a characteristic pathological feature of MS . Current treatments aim at preventing destruction of the myelin sheath by attenuating the autoimmune and inflammatory responses. However, there is a need for drugs that would directly prevent the process of demyelination and promote re-myelination, and current developments are in search of new drug targets to fulfil this goal .
In our study, AT2R stimulation prevented demyelination and promoted re-myelination. It was beyond the scope of the present study to provide ultimate answers about the mechanism of action resulting in AT2R-mediated protection of the myelin sheaths. Nevertheless, some data generated in the present study point to certain mechanisms involved.
Histological staining of spinal cord sections in our study clearly showed a co-localization of demyelinated areas with cell clustering. In C21-treated mice, this co-localization was still recognizable, but demyelination and cell clustering were both attenuated. FACS analysis of spinal cord tissue revealed that the amount of total and CD4+ T-cells in EAE mice was significantly reduced after pharmacological AT2R stimulation, suggesting that the clustered cells, which were co-localizing with demyelination and reduced in C21-treated mice, may have consisted of infiltrating inflammatory cells including T-cells. There are two potential reasons for the reduction in infiltrating T-cells, which are (i) inhibition of T-cell activation in the periphery, or (ii) protection of the blood–brain barrier (BBB) thus preventing T-cell infiltration into the brain.
Regarding T-cell activation in the periphery, FACS analysis of T-cells in lymph nodes/spleen of EAE mice did not reveal any difference between C21- and vehicle-treated animals arguing against an effect on peripheral T-cell activation. However, in studies performed by our group using another mouse autoimmune model, rheumatoid arthritis, there was an inhibition of peripheral T-cell activation in the spleen by AT2R stimulation . Therefore, the lack of the effect of C21 on T-cell activation in the periphery in the present study may be due to a non-optimal time point for this analysis, but does not generally exclude that an effect of C21 on peripheral T-cell activation contributed to the reduction in T-cell infiltration into the spinal cord.
Regarding the BBB, it has been shown that AngII via the AT1R impairs BBB integrity and that AT1R blockers exert protective effects [38,39]. With regard to the AT2R, one study indicates increased AT2R immunoreactivity on astrocytes in areas of BBB breakdown ; however, the selectivity of AT2R antibodies has been questioned . A recent publication reported reduced permeability of the BBB in response to C21 in a stroke model, but the mechanism resulting in the reduction of permeability was not studied . It was beyond the scope of the present study to study the effect of C21 on the BBB in detail. Nevertheless, a positive effect of AT2R stimulation on BBB integrity is conceivable according to the existing data and because therapeutic actions of AT1R blockers, which protect the BBB, are at least in part due to indirect stimulation of unopposed AT2Rs by reactively increased levels of AngII .
Our observation that AT2R stimulation reduced T-cell infiltration into the spinal cord in EAE and that cell infiltrates co-localized with areas of demyelination suggests that protection from demyelination was secondary to inhibition of the T-cell response. However, in aggregating rat brain cell cultures, which are devoid of T-cells, treatment with C21 still resulted in protection from IFNγ/LPS-induced demyelination. This observation points to additional mechanisms of action, by which AT2R stimulation protects the myelin sheaths.
In this context, another important finding of our study revealed by FACS analysis was a reduction in microglial activation in spinal cords from C21-treated mice in the acute phase of MOG-EAE. Furthermore, microglia in aggregating brain cell cultures stained with IB-4, a classical microglial marker, were found to be significantly reduced after C21-treatment when compared with vehicle treatment. Microglia have been described to contribute to MS/EAE by antigen presentation and by releasing pro-inflammatory cytokines and mediators, which in turn cause oligodendrocyte injury thus contributing to demyelination [5,43]. We could in fact show in primary microglia from rat hypothalamus and in a mouse microglia cell line challenged with LPS/IFNγ that AT2R stimulation significantly inhibited expression of the marker of microglia activation, CD11b, expression of pro-inflammatory cytokines (IL-1β and IL-6) and release of NO, suggesting that the cytotoxic effect of microglia in EAE is dampened.
It can be speculated that an AT2R-induced inhibition of the transcription factor nuclear factor κB (NF-κB), which has been shown by us and others [44,45], may be responsible for the reduced synthesis of cytokines. NF-κB is activated in microglia during EAE , and an inhibition of this transcription factor leads to an amelioration of the disease .
In contrast with our results demonstrating an inhibition of microglia activation by AT2R stimulation, McCarthy et al.  reported that AT2R stimulation caused an increase in the number of activated microglia in the core of ischaemic CNS damage coinciding with a reduction in infarct size, improved neurological outcome and increased neuronal survival. Very little is known so far about the impact of AT2Rs on microglia activation. Therefore, it is currently not possible to conclusively explain the different findings in both studies. However, in general, microglia can have disease-promoting as well as protective features depending, for example, on the type of injury, time course or microenvironmental conditions . They can polarize into a classical neurotoxic phenotype, known as M1, or they can differentiate into an alternative phenotype, known as M2, which is more neuroprotective . During disease, both of these opposite subdivisions as well as intermediate states may be present, therefore regulating the immunological process. In our study, the role of direct AT2R stimulation was evaluated in the classical M1 pro-inflammatory state, which can be induced by IFNγ, LPS or other Toll-like receptor (TLR) activators .
As discussed in the preceding sections, our data provide evidence that inhibition of the T-cell response and of microglia activation are mechanisms underlying the prevention of demyelination by AT2R stimulation. However, these findings do not exclude a direct protective effect of the AT2R on those cells responsible for synthesis of myelin components and of nourishing the myelin sheath, the oligodendrocytes, which indeed express AT2Rs [49,50]. Possible evidence for a direct effect on oligodendrocytes comes from our in vitro studies, in which AT2R stimulation not only prevented demyelination, but also accelerated re-myelination. Notably, re-myelination occurred when cells were no longer exposed to a pro-inflammatory stimulus. Thus any changes to the process of re-myelination occurring in our study were rather independent from effects on the inflammatory processes. Such an effect on oligodendrocytes could either consist in the protection of these cells from cell death and apoptosis or in an accelerated differentiation of OPCs into mature cells capable of synthesizing myelin components [34,51]. Therefore, further experiments are required in order to clarify the role of the AT2R in re-myelination, or more specifically, whether the AT2R is directly involved in the protection of OPCs from apoptosis or in the differentiation of OPCs into mature oligodendrocytes. A protective effect on oligodendrocytes may be facilitated by the neurotrophin BDNF acting on TrkB receptors as has been shown in several studies during the past 15 years [52–54]. A recent study from our group indeed found that AT2R stimulation leads to an increased synthesis of BDNF and TrkB in neuronal tissue in vitro and in vivo, and that the BDNF/TrkB axis seems essential for the neuroprotective actions of the AT2R . AT2R stimulation has also been shown very recently to induce BDNF synthesis in cerebromicrovascular endothelial cells . In concordance with our observation of AT2R-mediated re-myelination, Reinecke et al.  showed in crushed sciatic nerve injury that application of AngII significantly increased the degree of myelination and the axonal diameter, indicating re-myelination and axonal regeneration, respectively. This effect of AngII was abolished by the AT2R antagonist PD 123319, but not by the AT1R antagonist losartan, pointing to an AT2R-mediated effect.
Although the present study is the first to show a therapeutic effect of AT2R stimulation in a model of MS and–more generally–in an autoimmune model, further studies are needed to elucidate the related mechanisms of action in more detail.
While the present study found beneficial effects of AT2R stimulation on myelination, T-cell infiltration and microglia activation, it did not provide final answers to the mechanism underlying reduced T-cell infiltration, which may be an effect on T-cell activation in the periphery or a protection of the BBB or both. Future studies will also have to clarify the mechanism responsible for protection from demyelination and promotion of re-myelination, in particular whether direct effects on oligodendrocyte apoptosis and differentiation are involved, or whether the effects seen are secondary to attenuation of CNS inflammation.
It is a further limitation that the AT2R specificity of C21 effects was not proven in vivo. However, we provide in vitro evidence that all major protective effects of C21 were AT2R-mediated and could be inhibited by an AT2R antagonist. Furthermore, several in vivo studies by others and us have shown that the doses for C21 used in our study are AT2R-specific in vivo [30,55,56].
Finally, in the present study a positive treatment effect of C21 in the EAE model was only shown for a preventive protocol, i.e. with a start of treatment 3 days before immunization. Start of treatment at the time of onset of symptoms would have been of more clinical relevance. However, the present study still showed that, in principle, AT2R stimulation is able to interfere with the pathological events in EAE. The reason for only performing a prevention protocol is that the AT2R agonist used in this study, C21, which is the only metabolically stable AT2R agonist currently available and therefore the only one suitable for an in vivo study of longer duration, does not or only very poorly cross the BBB . Thus, C21 is still suitable for protocols with preventive treatment, since in EAE the BBB is being opened twice intentionally by pertussis toxin to facilitate penetration of the antigen into the CNS. Moreover, ‘natural’ BBB breakdown occurs mainly in very early stages of EAE and MS . Therefore, for protocols with start of treatment at onset of symptoms and–more importantly–for clinical development of AT2R agonists for treatment of MS, AT2R-agonistic molecules with better BBB penetrance are needed. Such molecules are currently indeed being developed in collaboration with the Department of Medicinal Chemistry at Uppsala University, which designed and synthesized C21 , and our group. That project has been supported by the American Multiple Sclerosis Society (Fast Forward Program: http://www.nationalmssociety.org/fast-forward/index.aspx).
Verónica Valero-Esquitino, Pawel Namsolleck, Florianne Monnet-Tschudi, Christine Brandt, Colin Sumners, Christa Thoene-Reineke, Thomas Unger and Muscha Steckelings designed the research; Verónica Valero-Esquitino, Kristin Lucht, Pawel Namsolleck, Florianne Monnet-Tschudi, Tobias Stubbe, Franziska Lucht, Friederike Ebner, Daniel Villela, Leon Danyel and Meng Liu performed the research; Verónica Valero-Esquitino, Kristin Lucht, Pawel Namsolleck, Florianne Monnet-Tschudi, Christine Brandt, Ludovit Paulis, Björn Dahlöf, Colin Sumners, Anders Hallberg, Thomas Unger and Muscha Steckelings analysed the data; and Verónica Valero-Esquitino, Colin Sumners and Muscha Steckelings wrote and revised the paper.
We thank Brigitte Delacuisine and Denise Tavel for excellent technical assistance.
The present study was funded by a fellowship from the Berlin-Padua-Gdansk PhD programme (to V.V.E.) and in part by a research grant from the Swiss Society for Multiple Sclerosis.
angiotensin type 1 receptor
angiotensin type 2 receptor
brain-derived neurotrophic factor
complete Freund's adjuvant
central nervous system
Dulbecco's modified Eagle's medium
experimental autoimmune encephalomyelitis
myelin oligodendrocyte glycoprotein
myelin oligodendrocyte glycoprotein-induced experimental autoimmune encephalomyelitis
nuclear factor κB
oligodendrocyte precursor cell
tumour necrosis factor