Fingolimod effects in neuroinflammation: Regulation of astroglial glutamate transporters?

Abstract

Fingolimod is an oral sphingosine-1-phosphate-receptor modulator which reduces the recirculation of immune cells and may also directly target glial cells. Here we investigate effects of fingolimod on expression of astroglial glutamate transporters under pro-inflammatory conditions. In astrocyte cell culture, the addition of pro-inflammatory cytokines led to a significant downregulation of glutamate transporters glutamate transporter-1 (slc1a2/SLC1A2) and glutamate aspartate transporter (slc1a3/SLC1A3) expression on the mRNA or protein level. In this setting, the direct application of fingolimod-1 phosphate (F1P) on astrocytes did not change expression levels of slc1a2 and slc1a3 mRNA. The analysis of both transporters on the protein level by Western Blot and immunocytochemistry did also not reveal any effect of F1P. On a functional level, the addition of conditioned supernatants from F1P treated astrocytes to neuronal cell culture did not result in increased neurite growth. In experimental autoimmune encephalomyelitis as a model of multiple sclerosis, fingolimod treatment reduced T cell and macrophages/microglia mediated inflammation and also diminished astrocyte activation. At the same time, fingolimod restored the reduced expression of slc1a2 and slc1a3 in the inflamed spinal cord on the mRNA level and of SLC1A2 and SLC1A3 on the protein level, presumably via indirect, anti-inflammatory mechanisms. These findings provide further evidence for a predominantly peripheral effect of the compound in neuroinflammation.

Fig 1. Effects of fingolimod-1 phosphate on glutamate transporter mRNA and protein levels in astrocyte cell culture.

RT-PCR analysis of (A) slc1a2 expression or (B) slc1a3 expression after astrocyte culture for 12 days and 48 hours of stimulation with TNF-α (100 U/ml) and IL-1β (10 ng/ml) versus naïve controls and with our without addition of F1P at 100 nM, n = 17/3/9/14 per group. (C,D) Western Blot analysis of (C) SLC1A2 protein expression or (D) SLC1A3 protein expression in astrocytes cultured for 12 days either as naïve cells or after stimulation with 100 U/ml TNF-α and 10 ng/ml IL-1β for 48 hours with or without F1P at 100 nM. Densitometric analysis of SLC1A2 or SLC1A3 protein levels was performed in relation to glyceraldehyde 3-phosphate dehydrogenase (GAPDH) as loading control, n = 3–4 per group, data are pooled from two experiments. Arrows indicate specific band. (E,F) Confocal imaging after immuncytochemistry for GFAP (red) and SLC1A2 or SLC1A3 (green) in inflamed astrocyte culture (stimulation with TNF-α and IL-1β) with or without addition of 100nM F1P for 48 hours. Images from representative cultures are shown. Bar = 20 µm for G,H. Data are given as mean ± SD for RT-PCR data or mean ± SEM for Western Blotting. * p < 0.05 or *** p < 0.01, t-test or Kruskal Wallis-test.

Fig 2. Clinical course and blinded histpathological analysis of inflammation and degeneration after treatment of MOG-EAE with fingolimod at 3 mg/kg/d.

(A) Clinical courseo f EAE after prophylactic treatment starting from day 0 after immunization (n = 6 per group). (B) Treatment initiation at the beginning of disease (day 11 p.i., n = 6 per group). (C) Treatment initiation at the early chronic phase of the disease (day 25 p.i., n = 8 per group). (D-G) Blinded histopathological quantification of spinal cord cross sections after staining for (D) CD3 positive T cells, (E) Mac-3 positive macrophages and microglia, (F) demyelination with the Luxol Fast Blue technique, and (G) axonal densities with Bielschowsky silver impregnation. In A-C, the day on the X axis indicates the start of treatment: directly after immunization (d0, n = 6 per group), at the first sign of symptoms (d11, n = 6 per group) or at the early chronic phase of the disease (d25, n = 8 per group). Experiments were analysed at the maximum of disease (days 15 or 17 p.i., respectively) for treatment start on day 0 and 11 p.i. and in the later chronic phase of EAE (day 80 p.i.) after treatment start on day 25 p.i. All data are given as mean ± SEM. Arrows indicate start of treatment.* p < 0.05, *** p < 0.001, Mann-Whitney test.

Fig 3. Effects of fingolimod on astrocyte activation and glutamate transporter mRNA levels in the spinal cord during EAE.

(A, B) Representative GFAP staining of spinal cord cross sections from a mouse treated with fingolimod in a preventive setting and a sham treated control at the maximum of EAE. Note the reduced GFAP staining after fingolimod treatment. Bar = 100µm (C) Blinded quantification of GFAP immunreactivity as marker of astrocyte activation on spinal cord cross sections. (D) RT-PCR analysis of slc1a2 expression. (E) RT-PCR analysis of slc1a3 expression, n = 6–8 per group. The mRNA expression of a prophylactically treated mouse is set to 1 as reference. All data are given as mean ± SEM. Please note that data are compiled from separate experiments with different starting points of fingolimod application (d0, 11 or 25). The day (d) on the X axis indicates the start of treatment directly after immunization (d0, n = 6 mice per group), at the first sign of symptoms (d11, n = 6 mice per group) or at the early chronic phase of the disease (d25, n = 8 mice per group). Experiments were analysed at the maximum of disease (days 15 or 17 p.i., respectively) for treatment start on day 0 and 11 p.i. and in the later chronic phase of EAE (day 80 p.i.) after treatment start on day 25 p.i. * p < 0.05, ** p < 0.01; *** p < 0.001, Mann Whitney test.

Fig 4. No effects of fingolimod on spinal cord glutamate transporter protein levels during EAE.

(A-D) Western Blot analyses of spinal cord homogenates relative to GAPDH (A) or beta actin (C) with the respective densitometry relative to the housekeeping gene (B,D). At the maximum of EAE, there was a decrease for SLC1A2 (A,B) and SLC1A3 protein levels (C,D) as compared to naïve mice which was restored after fingolimod treatment (3 mg/kg once daily). One out of three experiments is shown, n = 3 per group. (E-H) Laser scanning microscopy of spinal cord cross sections after staining for GFAP (red) and SLC1A2 (green, E,F) or SLC1A3 (green, G,H). Arrows indicate double labelled profiles. Representative images of spinal cord cross sections are shown. Bar = 50 μm.

Fig 5. Fingolimod conditioned astrocyte supernatants do not exert growth promoting effects in PC 12 cells.

(A-D) Representative images of PC12 cells in culture after hematoxylin eosin staining. Insets in (A) and (B) show higher manigifcation with representative neurite growth indicated by arrows. Bar indicates 50 µm in D and 20 µm in inset. As compared to (A) medium only as negative control and (B) the addition of BDNF as positive control, the addition of conditioned supernatnats from IL-1β and TNFα inflamed astrocytes with or without F1P treatment at 100 nM (C,D) did not lead to increased neurite lenght. (E) Blinded quantification of neurite lenghts in PC 12 cell culture. ACM, astrocyte conditioned supernatant. Data are given as mean ± SEM, n = 3 per group, 1 out of 2 experiments is shown. * p < 0.05 for medium versus addition of BDNF as positive control, Kruskal Wallis test.