A TSPO ligand is protective in a mouse model of multiple sclerosis

Abstract

Local production of neurosteroids such as progesterone and allopregnanolone confers neuroprotection in central nervous system (CNS) inflammatory diseases. The mitochondrial translocator protein (TSPO) performs a rate-limiting step in the conversion of cholesterol to pregnenolone and its steroid derivatives. Previous studies have shown that TSPO is upregulated in microglia and astroglia during neural inflammation, and radiolabelled TSPO ligands such as PK11195 have been used to image and localize injury in the CNS. Recent studies have shown that modulating TSPO activity with pharmacological ligands such as etifoxine can initiate the production of neurosteroids locally in the injured CNS. In this study, we examined the effects of etifoxine, a clinically available anxiolytic drug, in the development and progression of mouse experimental autoimmune encephalomyelitis (EAE), an experimental model for multiple sclerosis (MS). Our results showed that etifoxine attenuated EAE severity when administered before the development of clinical signs and also improved symptomatic recovery when administered at the peak of the disease. In both cases, recovery was correlated with diminished inflammatory pathology in the lumbar spinal cord. Modulation of TSPO activity by etifoxine led to less peripheral immune cell infiltration of the spinal cord, and increased oligodendroglial regeneration after inflammatory demyelination in EAE. Our results suggest that a TSPO ligand, e.g. etifoxine, could be a potential new therapeutic option for MS with benefits that could be comparable to the administration of systemic steroids but potentially avoiding the detrimental side effects of long-term direct use of steroids.

Figure 1. Expression of TSPO in the spinal cord of normal and EAE mice

1. TSPO was generally expressed at low levels in the normal adult spinal cord.

2. In contrast, TSPO was markedly upregulated in EAE mice.

3. The main site of TSPO in healthy mice was the central canal ependymal cells, which remained in similar high levels of TSPO expression after EAE.

4. TSPO was also expressed in a small population of NG2⁺ cells in the normal state and remained similar in the disease state.

5. TSPO was observed in activated astrocytes (GFAP⁺) in EAE mice.

6. The majority of the cells expressing TSPO were activated microglia and infiltrating macrophages (CD68⁺) in EAE mice.

7. TSPO mRNA levels also showed a dramatic increase in response to inflammatory signals. EAE spinal cords (*p = 0.004), as well as astrocytes (*p = 0.006) and microglia (*p = 0.01) treated with IFN-γ all showed an increase in TSPO mRNA levels. The p-values were calculated by t-test, n = 8/group. Scale bar = 50 µm.

Figure 2. Effect of etifoxine treatment starting at day 7 p.i. in EAE mice

1. A. Etifoxine treatment was started on day 7 post induction, and was administered daily i.p. at a dose of 50 mg/kg.

2. B. The drug treated group showed a decrease in clinical scoring at days 14–16.

3. C. The drug treated group showed a significant increase in weight retention during the peak of clinical symptoms.

4. D–G. The mice showed a significant decrease in the total scoring (*p = 0.000001) (D), an increase in the survival rate (E), a decrease in the maximal score (*p = 0.03) (F), and a delayed onset of clinical scoring (*p = 0.04) (G). The p-values for EAE clinical scores and weight were calculated using Mann–Whitney test, for total scores using Wilcoxon signed-rank test, and for days till onset and maximal score using t-test, n = 10 mice/group.

Figure 3. Histological and mRNA analysis of the inflammatory cytokines in the vehicle- or etifoxine-treated mice at onset of clinical symptoms

1. A–D. At day 10 p.i., drug treated animals showed significant differences in MBP (A), TSPO (B), Iba1 (C) and CD4 staining (D). (A) Animals treated with etifoxine showed an increase in retention of percentage of MBP coverage (*p = 0.001). Etifoxine treated animals also exhibited a decrease in TSPO⁺ cells per mm² (*p = 0.004) (B), along with less Iba1⁺ cells per mm² (*p = 0.02) (C). (D) The vehicle group showed infiltration of CD4⁺ cells, while no CD4⁺ cells were seen in the drug treated group. Infiltrated CD4⁺ cells were positive for TSPO expression.

2. E. The mRNA levels corroborated the CD4 histology. Along with the decrease in CD4 mRNA (*p = 0.01), there was a decrease in transcripts of the inflammatory cytokines interferon-γ (IFN-γ) (*p = 0.02) and interleukin 17 (IL-17) (*p = 0.04) in animals treated with etifoxine. The p-values were calculated by t-test, n = 8/group.

Figure 4. Effect of etifoxine treatment starting at day 18 p.i. in EAE mice

1. A–C. Treatment of etifoxine was started at the peak of EAE (day 18), and continued daily at a dose of 50 mg/kg i.p. (A) The drug treated group showed a decrease in clinical scores (B), with an increase in weight gain (C).

2. D–G. The etifoxine group also showed a decrease in the total scoring (D), greater animal survival (E), a decrease in the mean low score (*p = 0.01) (F), and a decrease in days of paralysis after treatment (*p = 0.004) (G). The p-values for EAE clinical scores and weight were calculated using Mann–Whitney test, for total scores using Wilcoxon signed-rank test, and for the mean low score and days of paralysis after treatment using t-test, n = 12 mice/group.

Figure 5. Histological analysis of the vehicle- or etifoxine-treated mice at the recovery phase of EAE

At the peak of EAE clinical symptoms, the drug treated group retained more MBP⁺ total percentage area than the control (*p = 0.02), along with less Iba1⁺ cells per mm² (*p = 0.004) and less infiltrating CD4⁺ cells per mm² (*p = 0.03). The p-values were calculated using t-test, n = 8/group.

Figure 6. Etifoxine modulation of T-cell activity during EAE

Spinal cords of vehicle- or etifoxine-treated mice were either used for flow cytometry analysis or ELISA analysis. Cells stained for CD4, CD8, IL-17 and IFN-γ were gated according to fluorescent intensity.

1. Cells were then measured as a percentage of the total cell population. There was a decrease in the percentage of CD4⁺ (*p = 0.04), CD4⁺/IL-17⁺ (*p = 0.04), CD8⁺ (*p = 0.01) and CD8⁺/IFNγ⁺ (*p = 0.03) cells in the drug-treated group

2. Protein levels of the inflammatory cytokines IL-1β (*p = 0.001), IL-17 (*p = 0.02) and IFN-γ (*p = 0.02) were also measured, and there was a significant decrease in all cytokines in response to etifoxine treatment

3. The p-values were calculated using t-test, n = 8/group.

Figure 7. Effect of etifoxine treatment on NG2⁺ cells in the spinal cord during the recovery phase of EAE

1. There was a significant increase in the NG2⁺ cell population in the etifoxine treated group (*p = 0.01) compared to the vehicle treated group, and there was also an increase in Olig2⁺ cells.

2. NG2⁺ cells were found on the dorsal side of the central canal, and extended toward sites of injury, and co-labelled with TSPO.

3. The mRNA levels of Plp (*p = 0.007) and Olig2 (*p = 0.02) were increased in the drug treated group, so were the levels of akr1c1 (*p = 0.03), the main 3α-hydroxysteroid dehydrogenase (3αHSD) that is responsible for the biosynthesis of the neurosteroid allopregnanolone. The p-values were calculated using t-test, n = 8/group. Scale bar = 50 µm.

Figure 8. Electron microscopy of the lumbar section of the spinal cord in EAE mice after vehicle or etifoxine treatment

1. Low magnification electron micrograph showed a large number of degenerating myelinated axons (dAx) and microglia/macrophage cells (M) in the spinal cord white matter of the vehicle-treated EAE mice. Note that the degenerating axons displayed various morphological features, but they all showed disorganized myelin sheaths and electron-dense or swollen axoplasm. Microglia/macrophage cells (M) were frequently found to surround or be close to the degenerating axons. Scale bar = 5 µm.

2. High magnification electron micrograph taken from the same section in panel A, demonstrating that several microglia cells surrounding a degenerating axon (dAx) where the myelin sheath completely collapsed and the axoplasm became transparent, and indicating few organelles existed at this advanced stage of degeneration. Another degenerating axon (dAx) was seen near the microglia/macrophage (M). Scale bar = 1 µm.

3. Low magnification electron micrograph showing the spinal cord white matter of etifoxine-treated mice after EAE. Most of the axons (Ax) were myelinated and axonal organelles such as mitochrondria were present. No microglia or macrophages were found associating with myelinated axons. Scale bar = 5 µm.

4. High magnification electron micrograph showing myelinated axons (Ax), while mitochrondria appearing normal and neurofilaments being well preserved in the axoplasm. Inset showing higher magnification image of a myelinated axon (asterisk), with myelin sheath layers being clearly identified, and the major dense lines and the intraperiod lines also being clearly seen. Scale bar in D = 1 µm and in the inset = 0.1 µm.