2-BFI Attenuates Experimental Autoimmune Encephalomyelitis in Mice via Regulation of Lymphocytes Subsets

Background: Imidazoline compounds are well accepted to exhibit various pharmacological effects including antidepressant, anti-inammatory, analgesic, anti-morphine tolerance and inhibit the activity of monoamine oxidase. 2-(-2-benzofuranyl)-2-imidazoline (2-BFI), a selective imidazoline 2 receptor (I2R) ligand, has been proven to exhibit therapeutic effects for various neuroimmunological diseases. However, the mechanism behind its neuroprotective properties remains elusive. Methods: In this study, we used 2-BFI for the treatment of mice with experimental autoimmune encephalomyelitis (EAE) induced by myelin oligodendrocyte glycoprotein (MOG33-55). The clinical signs of neurological decits were evaluated daily. The demyelination and inammatory inltration in the CNS of mice with EAE were examined by Luxol Fast Blue (LFB) staining and hematoxylin-eosin (H&E) staining. Flow cytometry was utilized to examine the ratios of lymphocyte subsets in the periphery and CNS of mice with EAE. We also used Reverse Transcription–Polymerase Chain Reaction (RT-PCR) to observe the changes of expression of inammatory cytokines by 2-BFI intervention. Results: We found that 2-BFI signicantly reduced the incidence of EAE and attenuated the severity of neurological disability. Pathological staining showed that the inltration of inammatory cells and demyelination in the central nervous system (CNS) of the mice were markedly alleviated via 2-BFI intervention. To explore the mechanism of action of 2-BFI, we used ow cytometry to determine immunophenotypes in the spleen and CNS of the mice. We discovered that 2-BFI signicantly decreased the ratio of CD28+ lymphocytes and B cells in the spleen of EAE mice. In the CNS, the expression of CD4+ T cells was downregulated by 2-BFI, while B cells and CD39+ lymphocytes


Background
Imidazoline receptors (IR) are widely distributed in mammalian tissues including the central nervous system (CNS) and are generally classi ed into three subtypes: I1R, I2R and I3R [1]. I2R are predominantly located in the brain and liver where they regulate monoamine turnover identi ed by idazoxan [2]. A recent study revealed that 2-(benzofuranyl)-2-imidazoline (2-BFI) is a speci c ligand to I2R [3]. In the CNS, I2R binding is thought to mainly occur in the outer membrane of the mitochondria of neurons and astrocytes [4,5]. There are enormous evidences that I2R are associated with various neurological disorders, and its ligands, including 2-BFI, manifested protective effects against these diseases. Prolonged treatment with the imidazole compound LSL 60101 could activate I2R and protect motoneurons from death caused by neurectomy of the facial motor nerve [6]. Selective I2R agonist reduced the size of infarct in rats with focal ischemia, and improved neurobehavioral de cits [7]. Chronical administration of 2-BFI can reduce the development of tolerance to morphine [8], and showed therapeutic potentials in psychiatric disorders [9]. Moreover, advances in the eld of neuroin ammation have indicated that 2-BFI can suppress the accumulation of oxidants, inhibit the in ammatory response and enhance neural viability [10]. In a study on neuropathic pain model in rats, 2-BFI showed anti-hypersensitivity and anti-in ammatory effects by downregulating the levels of proin ammatory cytokines [11]. Thus, it is crucial to explore the mechanism underlying the neuroprotection neuroprotective properties of 2-BFI to con rm its potential value in clinical application.
Multiple sclerosis (MS), a prototypical autoimmune in ammatory disorder of the CNS characterized by demyelination and in ammatory in ltration, is believed to be mediated by autoreactive T cells.
Experimental autoimmune encephalomyelitis (EAE) is a well-established animal model of MS, which mimics the clinical features and pathological manifestations found in this disease [12]. The migration of autoreactive myelin-speci c T cells across the blood-brain barrier (BBB) leads to a diverse accumulation of T cells, B cells and macrophages in the CNS, which is considered to be a crucial step in the initiation of in ammation and progressive paralysis in EAE [13,14]. Recent evidence suggests that perturbation in lymphocyte subsets including CD4 + T cells, CD8 + T cells and B cells is a signi cant hallmark in the progression of EAE. Meanwhile, extensive research con rms that cytokines derived from T cells such as interferon-γ (IFN-γ) play a key role in the development of EAE [15][16][17], whereas B cells ameliorate the severity of the disease via the production of suppressive cytokines like interleukin-10 (IL-10) and TGF-β [18,19].
In our previous studies, 2-BFI was found to signi cantly attenuate spinal cord injury and exhibited neuroprotection against EAE [20]. We also found that 2-BFI could regulate the expression of cytokines in EAE with dose and time dependency [21]. It is reasonable to deduce that 2-BFI serves important functions in in ammatory and immune responses of EAE, but the mechanism has not been fully elucidated. In this study, 2-BFI was administrated to mice with EAE induced by myelin oligodendrocyte glycoprotein (MOG33-55). We evaluated the therapeutic effect of 2-BFI against the clinical neurological de cit and pathological damage. We used ow cytometry to detect the in uence of 2-BFI on the changes of lymphocyte subsets in the periphery and CNS. We further examined the changes of mRNA levels of proin ammatory cytokines such as IFN-γ, IL-10 and TGF-β to explore the potential immunological protective mechanisms of 2-BFI.

Mice and EAE induction
Female C57BL/6 mice, 6-8 weeks old, weighing 17-20 g were obtained from the Experimental Animal Center of Beijing Military Region and housed under speci c pathogen-free conditions. Animals were maintained with free access to food and water. Each mouse received subcutaneously injections into multiple sites in the anks with 0.2 mg of MOG35-55 emulsi ed in IFA supplemented with 8 mg/ml of heat-killed mycobacterium tuberculosis H37Ra (complete Freund's adjuvant, CFA). In addition, 0.2 ml phosphate-buffered saline (PBS) containing 200 ng of PT were injected intraperitoneally (i.p.) at days 0 and 2 post-immunization. The body weight and clinical signs of EAE micewere examined every day.

2-BFI treatment and treatment groups
Twenty-four mice were randomly divided into three groups for the present experiments. EAE mice were subtyped into one 2-BFI-treated groups (EAE-2BFI group) and one saline-treated group (EAE-control group,). Mice in the EAE-2BFI group were given twice daily i.p. injection of 2BFI 20 mg /kg for 14 days from the day of immunization, while the EAE-control group mice were given twice daily i.p. injection of the same dose of saline. The remaining mice (CFA-control group) were immunized with only CFA and the equivalent amount of saline instead of MOG35-55 used in EAE mice.

Clinical and pathological evaluation of EAE
Clinical signs of neurological de cits were observed and evaluated daily, the severity was graded as follows scale [22]: (1) for the tail, 0 = no signs; 1 = half paralyzed tail; 2 = complete limp tail; (2) for each of the hind-or forelimbs assessed separately, 0 = no signs; 1 = a weak or altered gait; 2 = paresis; 3 = a fully paralyzed limb. The sum of the state of the tail and all four limbs was from 0 to 15, thus a fully paralyzed quadriplegic mouse would attain a score of 14. Mortality equals a score of 15.
EAE mice and CFA-control mice were euthanized at 20 days after immunization, approximately the peak clinical signs exhibited. The mice were perfused by intracardiac injection pre-cooling PBS, then the lumbar cords were dissected out and xated in 4% paraformaldehyde for 12h. Transverse sections of the spinal cord were stained with H&E staining. The severity of in ammatory cell in ltration was assessed according to the following scoring system described by Okuda et al. [23]: 0 = no in ltration; 1 = slightly cellular in ltrates were around meninges; 2 = mild cellular in ltrates in parenchyma (1-10/section); 3 = cellular in ltrates in parenchyma (11-100/section); and 4 = severe cellular in ltrates in parenchyma (>100/section). The cerebral cortices of mice were sectioned and examined by LFB staining to evaluate the severity of demyelination. The LFB color intensities were recorded by Image J system and normalized to CFA-control group.
Isolation of mononuclear cells from the CNS and splenocytes Fresh brain and spinal cord tissues were removed from mice and rinsed in ice-cold PBS. The brain and remaining spinal cord were cut into small pieces, using the back of a sterile 1 ml syringe plunger press each piece of organ ltered through a 100mm wire mesh in a 10cm petri dish containing 10ml of ice-cold PBS. The tissues were dispersed into a single cell suspension and in ammatory cells were recovered from the CNS, then centrifuged at 1, 500 rpm for 5 min at 4°C to resuspend the cells in 30% Percoll and then layered over 70% Percoll. The cells were collected from the 30%/70% interface after centrifugation at 600 g for 20 min at room temperature and washed twice with PBS. Spleens were removed from the same mice and placed in PBS. Tissues were forced through 100-mesh stainless steel screens to give a single cell suspension. Red blood cells in spleen cell preparations were lysed by red blood cell lysis buffer, and the cells were washed and resuspended in PBS. The total number of cells derived from each mouse was counted by a hemocytometer for further staining.

Fluorescence-activated cell sorting analysis
Spleen or CNS-derived in ammatory cells (0.5-1 × 106) were washed with PBS, then incubated for 30 min at room temperature in the dark with speci ed antibodies such as anti-CD4-FITC, anti-CD3-APC, anti-CD8-PE or anti-B220-PE-cy5.5. Nonspeci c staining was determined by incubating cells with labeled isotypematched. Staining kit were purchased from Thermo Fisher Scienti c Inc. and used according to the manufacturer's instructions. Data collection was performed using a uorescence-activated cell sorting (FACS) Calibur ow cytometer (BD Biosciences, USA), and analyzed with Cell Quest.

Semi-quantitative RT-PCR
Total RNA was isolated from the brains of the mice with Trizol (Invitrogen, USA) in accordance with the manufacturer's guidelines. The total RNAs were reverse-transcribed (RT) to cDNA using a Gene Amp RNA PCR kit with oligo dT primer. PCR ampli cation of each cDNA target (IFN-γ, IL-10 and TGF-β) was performed from the same RT reaction according to the manual: Ampli cation conditions including annealing temperatures, number of cycles, and extension times were optimized for each target. PCR products were run on a 1.5% agarose gel containing 0.5 μg/ml ethidium bromide, and were visualized under UV light. The density of the band was quantitated using a Digital Imaging System.

Statistics analysis
Data were analyzed by the SPSS statistical program. Quantitative results were expressed as mean ±standard deviation. Data comparison between two groups was analyzed by Student's unpaired 2-tailed t test. One-way ANOVA was utilized for multiple group comparison followed by post hoc analysis with Dunnett's test to identify signi cant groups. A P value less than 0.05 was taken as a signi cant difference for all statistical analyses.

2-BFI attenuated neurological de cits of EAE mice
In this study, immunization with MOG35-55 induced 100% mice (8/8)  We noticed that mice in the EAE-control group developed ascending accid paralysis with the appearance of a accid tail around 11 to 13 d.p.i, and quickly deteriorated to hind leg paralysis and eventually complete paralysis. EAE-suffering mice display less activity, less feeding behavior as clinical signs became severe accompanied with the loss of body weight. The mobility de cits and severity of mice in the EAE-2BFI group were ameliorated, and the mean clinical score was signi cantly downregulated from 15 d.p.i to the day they were euthanized (Fig. 1). These data con rmed that 2-BFI improved the progression of EAE by attenuation of neurological de cits of mice.
2-BFI ameliorated the in ltration of in ammatory cells and demyelination in the CNS of EAE mice We further used H&E staining to evaluate the in ammatory cell in ltration in the spinal cord of the mice. In the CFA (complete Freund's adjuvant)-control group, no detectable in ammatory cell in ltration showed in the tissues. Meanwhile, serious in ammatory cell in ammation and perivascular cu ng were observed in the lumbar spinal cords, whereas the pathological changes of 2-BFI-treated mice were signi cantly alleviated with mild in ammatory invasion observed under the spinal meninges ( Fig. 2A). To evaluate the severity of demyelination, LFB staining was adopted to detect demyelination in the cerebral cortex sections of the mice. As expected, the EAE-control group showed typical signs of loose myelin while demyelination was reduced in the EAE-2BFI group based on LFB staining when compared with the EAEcontrol group (Fig. 2B). Consistently, the average pathological scores were markedly reduced in the EAE-2BFI group (1.75 ± 0.96 vs 3.25 ± 0.89, P = 0.022, Fig. 2C). The color intensities of LFB were elevated in the EAE-2BFI group compared those of the with EAE-control group (0.75 ± 0.13 vs 0.38 ± 0.06 normalized to the CFA-control group, P < 0.001, Fig. 2D). The amelioration and reduction of pathological lesions were consistent with the attenuation of clinical morbidity with 2-BFI treatment.

2-BFI regulated lymphocyte subsets in the spleen and CNS of EAE mice
To explore the cell-mediated immunity mechanisms underlying the neuroprotective effects of 2-BFI, we used ow cytometry to examine the expression of lymphocyte subsets in the periphery and CNS. Spleen and CNS lymphocytes were isolated separately from the same immunized mice as described above. As shown in Fig. 3A, the splenic CD4+, CD8 + T cells and CD28 + lymphocytes were decreased both in the EAE-control group and EAE-2BFI group compared with the CFA-control group. Moreover, quanti cation analysis con rmed that the percentages of CD28 + lymphocytes in the spleen were reduced in the EAE-2BFI group compared with the EAE-control group (10.92 ± 6.04% vs 19.43 ± 7.1%, P = 0.022, Fig. 3C). In contrast to the T cells, in the spleen of EAE mice, the percentages of B cells and CD39 + lymphocytes were increased compared with the CFA-control group, while 2-BFI treatment dramatically downregulated the ratio of B cells compared with that of the EAE-control group (31.22 ± 10.05% vs 54.35 ± 10.6%, P = 0.007, Fig. 3C).
In the CNS, very few in ammatory lymphocytes were detectable in the CFA-control group. CD4+, CD8 + T cells and CD28 + lymphocytes of the CNS in ltrates were decreased with 2-BFI treatment (Fig. 3B), but only the level of CD4 + T cells showed statistical signi cance compared with the EAE-control group (10.56 ± 10.33% vs 25.67 ± 12.73%, P = 0.021 Fig. 3D). Correspondingly, 2BFI signi cantly up-regulated the ratio of activated cells in the B cells (74.04 ± 22.29% vs 45.01 ± 18.29%, P = 0.01, Fig. 3D) and CD39 + lymphocytes (77.79 ± 13.96% vs 62.69 ± 13.7%, P = 0.049, Fig. 3D) in the CNS of EAE mice. These results suggested that the regulation of lymphocyte subsets may be the key mechanism of the therapeutic effects of 2-BFI against EAE.

2-BFI inhibited proin ammatory cytokines expression while promoted anti-in ammatory cytokines in the CNS of EAE mice
As we had proven that 2-BFI could regulate the expression of lymphocyte subsets in the CNS of EAE mice, it was necessary to observe how the related in ammatory cytokines affected the immune responses. RT-PCR was employed to detect the mRNA expression of the proin ammatory and anti-in ammatory cytokines (including IFN-γ, IL-10 and TGF-β) in the brains of the mice. RNA extraction, cDNA synthesis, imaging and semi-quantitation were performed as described in the Methods (Fig. 4A). We found that the mRNA level of proin ammatory cytokine IFN-γ which was previously reported as the key in ammatory factor in EAE-2BFI group were dramatically decreased compared with the EAE-control group (1.23 ± 0.49 vs1.87 ± 0.27, P = 0.006). Meanwhile, 2-BFI treatment signi cantly enhanced the mRNA expression of IL-10 (1.36 ± 0.51 vs 0.87 ± 0.27, P = 0.028) and TGF-β (1.91 ± 0.55 vs 1.04 ± 0.43, P = 0.0034) compared with the EAE-control group (Fig. 4B). The imbalance of expression of proin ammatory and antiin ammatory cytokines was consistent with the lymphocyte subsets in the CNS of EAE mice regulated by 2-BFI.

EAE is mediated by the imbalance of lymphocytes
EAE is an animal model for T cell-mediated autoimmune disease, which mimics the pathogenesis of MS.
In MOG-induced EAE, activated T cells cross the BBB and bind to the antigen presenting cells, which further promotes the activation of B cells and macrophages by stimulating the secretion of a large number of chemokines and cytokines. Peripheral T cells, monocytes and B cells are recruited to access the brain parenchyma, and result in in ammation and demyelination of the CNS [14]. Current studies have suggested that the migration of autoreactive myelin-speci c T cells from peripheral blood into the CNS is an important initiating factor for the onset of EAE. After peripheral T cells were activated, only a small number of lymphocytes crossed the BBB and in ltrated the CNS, causing mild neurological de cit.
With the enhanced permeability of the BBB, the CNS was invaded by millions of pro-in ammatory lymphocytes and macrophages in the second wave within hours, which was consistent with the clinical manifestations of MS, further suggesting that T cell migration plays a central role in the pathogenesis of MS [24,25]. In recent years, attention has been given to the migration of CD8 + T cells into the CNS during this process. It has been reported that in the pathogenesis of MS, CD8 + T cells reacted directly with demyelination and damage of axons by releasing granzyme B [26]. In EAE, the number of CD4 + T and CD8 + T cells that in ltrated the CNS were considered to be directly correlated with the progression and severity of EAE [27]. In the acute phase of EAE, B cell recruitment to the in ltrating areas of the CNS was also observed. Recent studies have demonstrated the functions of B cells in the immune system, including secreting immunoglobulin and cytokines, antigen presentation, and regulating effector T-cell differentiation [28]. It was generally considered that B cells play a key role in the regulation of EAE. In the process of interaction between T and B cells, self-reactive B cells act as e cient antigen presenting cells, and enhance Th2 polarization of MOG-speci c T cells rather than Th1 [29]. Studies con rmed that B cells regulate the recovery of EAE by secreting IL -10, TGF-β and IL-35, and block the activation of T cells and macrophages [30,31]. B cell de ciency is associated with numerous immunological abnormalities.
Therefore, B cells are considered to have potential therapeutic effects for a variety of autoimmune diseases [32].

2-BFI regulated the lymphocytes subsets in the CNS of mice with EAE
2-BFI, a high a nity ligand for I2R, is mainly distributed on the outer membrane of the mitochondrial of astrocytes, which are important components of the BBB [33]. In the early phase of EAE, it was noticed that activated astrocytes lose the perivascular end-feet, resulting in increased BBB permeability [34]. The appropriate dose of 2-BFI exhibits protective properties against astrocyte damage in an ischemic stroke [35,36]. It was reported that 2-BFI can enhance the intercellular tight junction via the downregulation of MMP-9 expression levels, and suppress vascular endothelial permeability to preserve cerebral microvascular endothelial integrity [37,38]. Our previous studies demonstrated that 2-BFI reduced the disruption of the BBB and alleviated oxidative stress in the CNS of EAE mice [20]. In this study, 2-BFI was proven to effectively reduce the incidence and alleviate the symptoms of neurological de cit in EAE mice. The in ltration of in ammatory cells and demyelination in the CNS of EAE mice were also ameliorated by 2-BFI treatment. Using ow cytometry to analyze the subsets of lymphocytes in the spleen and CNS of EAE mice, we found that the ratio of CD4 + T and CD8 + T cells in the spleen were reduced compared with that of the CFA-control group, and these proin ammatory lymphocytes were only detected in the CNS of the EAE mice, indicating that the MOG35-55 stimulated T cell migration from the periphery to the CNS during the pathogenesis of EAE. We further discovered that the percentage of CD4 + T cells in the CNS of EAE mice treated with 2BFI were signi cantly reduced than that in the EAE-control group, and the ratio of CD8 + T lymphocytes also tend to decreased, suggesting that 2-BFI inhibited the migration of in ammatory lymphocytes into the CNS. It is worth noting that 2BFI did not increase the CD4 + T and CD8 + T cells in the spleen. We want to further explore the effect of 2BFI on the proliferation and differentiation of peripheral T cells. CD28 is recognized as an important component of the B7/CD28 family, the classical costimulatory signal during naive T-cell activation [39,40]. It was con rmed that CD28 coordinated T cell proliferation, differentiation and survival, and was considered to play an key role in the progression of MS [41,42]. We found that with the intervention of 2-BFI, the peripheral CD28 expression of EAE mice was markedly downregulated, while the CD28 + lymphocytes in the CNS were not signi cantly decreased. These results indicate that 2-BFI may directly affect the proliferation and differentiation of peripheral in ammatory T cells and delay the progression of EAE by reducing the expression of CD28. Moreover, this study indicated that the percentage of B cells in the CNS of the EAE-2BFI group was signi cantly upregulated as well as obviously decreased in the spleen compared with those in the EAE-control group. The activation and migration of B cells depend on the stimulation signals provided by T cells. CD39 is mainly expressed on B cells, dendritic cells and Treg cells, and is considered to be a marker of activation of B cells [43]. We discovered that the ratio of CD39-labeled lymphocytes in the CNS of EAE mice was also markedly increased by 2-BFI intervention. Previous studies have con rmed the signi cant role of immune regulation of CD39 by clearing the pro-in ammatory stimuli [44]. Therefore, it is speculated that 2BFI promotes the migration of B cells to the CNS of EAE mice, increases the ratio of activated B cells and exerts immunomodulatory effects against EAE.
2-BFI regulated the expression of cytokines in the CNS of mice with EAE To validate the aforementioned results, we further examined the expression of cytokines in the CNS of the mice. We found that in the EAE-2BFI group, the level of IFN-γ, which is mainly produced by Th1 cells, decreased signi cantly, while the expression of anti-in ammatory cytokines such as IL-10 and TGF-β secreted by activated B cells dramatically increased. It has been reported that IL-10 inhibits the progression of EAE mediated by Th1 cells [45]. Evans et al. con rmed that B cells alleviate the in ammatory reaction by producing IL-10 to inhibit the differentiation of CD4 + T cell and the activity of effector T cell in animal models of collagen-induced arthritis [46]. A study of a mouse model of systemic lupus erythematosus suggested that anti-in ammatory cytokines secreted by B cells were found to inhibit the proliferation of T cell and the differentiation into Th1 cells [47]. In other animal models including type I diabetes, systemic lupus erythematosus, rheumatoid arthritis and contact dermatitis, B cells were con rmed to play negative regulatory roles during immune responses, and that the transfer of IL-10secreting B cells can effectively decrease in ammation [48]. It has also been reported that the antiin ammatory effect of regulatory B cells indirectly affects the proliferation of T cells by inhibiting the antigen presentation of dendritic cells and downregulating the expression of cytokines secreted by CD4 + T cells [30]. Our results further con rmed that 2BFI decreased the proliferation and activation of CD4 + T cells, and elevated the ratio of B cells in the CNS of EAE mice, suggesting that 2-BFI can regulate the lymphocyte subset rates to ameliorate the severity of the disease.

Conclusions
In this study, we demonstrated that the neuroprotective effect of 2-BFI against EAE was mainly achieved by inhibiting the activation and proliferation of pathogenic T lymphocytes and upregulating the ratio of B cells to reverse the immune imbalance induced by EAE. Further studies on I2R receptor ligands are expected to clarify the speci c mechanism and to provide a new idea in the treatment for neuroimmune diseases. Availability of data and material All data generated or analysed during this study are included in this manuscript.

Con ict of interests
The authors declare that the research was conducted in the absence of any commercial or nancial relationships that could be construed as a potential con ict of interest.

Authors' contributions
Xiaofeng Shang and Nana Xi analyzed the expression of lymphocytes subsets in the CNS and spleen of mice with EAE. Tan Wang performed the histological examination and was a major contributor in writing the manuscript. Xiaofeng Shang and Tan Wang contributed equally to this work. All authors read and approved the nal manuscript. Figure 1 2BFI treatment reduced the daily mean clinical score of EAE mice by attenuation of neurological de cits.

Figures
* indicates P < 0.05 compared with the EAE-control group. ** indicates P < 0.01 compared with the EAEcontrol group.   to CFA-control group. * means P < 0.05 compared with the EAE-control group.