Matrine protects retinal ganglion cells from apoptosis in experimental optic neuritis

Background: Inflammatory demyelination and axonal injury of the optic nerve are hallmarks of optic neuritis (ON), which often occurs in multiple sclerosis and is a major cause of blindness in young adults. Although a high dose of corticosteroids can promote visual recovery, it cannot prevent permanent neuronal damage. Novel and effective therapies are thus required. Given the recently defined capacity of matrine (MAT), a quinolizidine alkaloid derived from the herb Radix Sophorae flavescens, in immunomodulation and neuroprotection, we tested in this study the effect of matrine on ON in rats with experimental autoimmune encephalomyelitis, an animal model of multiple sclerosis. Results: MAT administration, started at disease onset, significantly suppressed optic nerve infiltration and demyelination, with reduced numbers of Iba1+ macrophages/microglia and CD4+ T cells, compared to those from vehicle-treated rats. Increased expression of neurofilaments, an axon marker, and decreased apoptosis in retinal ganglion cells (RGCs) were also observed after MAT treatment. Conclusions: Taken as a whole, our results demonstrate that MAT attenuated inflammation, demyelination and axonal loss in the optic nerve, and protected RGCs from inflammation-induced cell death. MAT may therefore have potential as a novel treatment for this disease that causes blindness. our observations, phosphorylated

apoptosis has been demonstrated after optic nerve injury (8). Once the optic nerve is damaged, RGCs will die and axons will fail to regenerate, leading to traumatic or ischemic nerve injury or degenerative conditions in the patient (9). The death of RGCs has been considered the main cause of vision loss after an episode of ON (10,11). In the animal model of relapsing/remitting EAE, RGC apoptosis begins within a few days after onset of optic nerve inflammation (12, 13), suggesting that axonal damage and cell loss are induced by optic nerve inflammation.
Matrine (MAT), a natural quinolizidine alkaloid compound extracted from the herb root of Sophorae flavescens, with a MW of 258.43 (C15H24N2O) (14)(15)(16), is known for its various effects in animal models of EAE, including protection against apoptosis, tumor and fibrotic tissue development, and inflammation (17). We have recently shown that MAT can ameliorate clinical signs and alleviate neuro-axonal injury in the CNS of EAE animals by regulatory T cells and reducing Th1 and Th17 cells in the CNS and periphery and increasing the number of neural protective molecules (18,19).
However, the ability of MAT to suppress ON and protect RGCs has not been studied.
In the present study we tested our hypothesis that MAT cannot only inhibit proinflammatory response, but also promote RGC survival by protecting these cells from inflammation-induced apoptosis. By using experimental ON in an EAE rat model, we examined the effect of MAT on inflammatory cell infiltration, demyelination, and neurodegeneration and RGC apoptosis of the optic nerve.

MAT treatment and disease assessment
Rats immunized with antigen emulsion were randomly divided into three groups (n = 10 each group): (1) MAT (MW: 264.36, a small molecule that was purchased from Chia-Tai Tianqing Pharmaceutical Co.), was dissolved in normal saline and injected intraperitoneally (i.p.) at 250 mg/kg daily, starting from day 11 after primary immunization (p.i.) until the end of the experiment (day 19 after p.i.); (2) immunized rats that received vehicle via i.p. served as control; (3) non-immunized naïve rats that received vehicle i.p. served as naïve control.
Rats were monitored and scored daily for clinical disease severity by two independent observers following the standard 0-5 EAE grading scale as previously published (6, 8): 0, natural; 0.5, partial tail paralysis; 1, tail limpness or waddling gait; 1.5, loss of tail tonicity or waddling gait; 2, hind limb weakness; 2.5, partial limb paralysis; 3, paralysis of one limb; 3.5, paralysis of one limb and partial paralysis of another; 4, paralysis of two limbs; 4.5, moribund state, and 5, death. Body weight was also recorded daily from the day of immunization.

Histopathological evaluation of optic nerves
To assess the extent of CNS inflammation and demyelination, the rats were sacrificed on 19 days after p.i.. Rats were anesthetized by intraperitoneal injection of 1% pentobarbital sodium (50 mg/kg) and extensive perfusion with 0.9% normal saline, the optic nerves and retinas were quickly removed and post-fixed with 4% paraformaldehyde. They were then embedded in paraffin, cut into paraffin sections (2-5 μm) and dewaxed in xylol, rehydrated. Sections were stained with hematoxylin and eosin (HE) for histological analysis and luxol fast blue (LFB) to detect demyelination. The histological examination was performed and scored by light microscopy by a blinded investigator using a grading scale as previously published criteria (Quinn, Dutt, and Shindler 2011): 0, no inflammatory infiltration; 1, a few scattered inflammatory cells of the optic nerve or optic nerve sheath; 2, moderate inflammatory infiltrates; 3, severe inflammatory infiltrates; 4, massive inflammatory infiltrates. Given that ON can occur either bilaterally or unilaterally, all subsequent analyses were performed using both eyes from each rat as individual data points. Scores of demyelination and inflammation were calculated by Image-Pro Plus 5.0 (IPP5.0) software.

Immunofluorescence analysis of optic nerves and retina cross-sections
The optic nerves and retinas from each group were paraffin-embedded and cut into 5-μm-thick sections for immunofluorescence. First, nonspecific binding was blocked with 3% bovine serum

Statistical analysis
All the animal groups were coded and analyses were conducted by two researchers blind to experimental conditions. SPSS 21.0 (SPSS, IBM, USA) and GraphPad Prism 5.0 (La Jolla, CA, USA) were used for the statistical analyses. Multiple comparisons were performed using the Kruskal-Wallis test, or ANOVA, followed by the LSD-t-test. Data are presented as mean ± SD, and p<0.05 was considered significant.

MAT treatment alleviates ongoing EAE in Wistar rats
To observe the therapeutic effects of MAT in ongoing EAE, body weight and clinical scores were measured, and disease severity was measured using the standard 0-5 grading scale. As shown in Fig.   1, clinical signs of EAE began on day 11 after p.i., when the treatment was started. When compared with the vehicle-treated rats, the clinical score of MAT-treated rats was significantly reduced (p< 0.05), confirming the effect of MAT treatment in ongoing EAE.

MAT treatment reduces optic nerve inflammation
In order to determine whether MAT ameliorated the degree of optic nerve inflammation, rats were euthanized on day 19 after p.i., optic nerve sections were examined by hematoxylin-eosin (HE) staining for areas of inflammatory cell infiltration, and inflammation was graded on a 0-to 4-point scale. Optic nerve sections from naïve rats served as normal controls. Consistent with the clinical scores, massive inflammatory infiltration was found in the optic nerve of vehicle-treated rats, while this infiltration was significantly decreased by MAT treatment (Fig. 2, P< 0.05).
To further confirm the infiltration of inflammatory cells, sections of optic nerve were stained with a macrophage/microglia-specific marker, Iba1 (Fig. 3A), and CD4 T cells (Fig. 3B), two important immune cell types in ON (9). Our results showed that the number of Iba1 + cells was largely increased in immunized rats; MAT treated rats had a significantly reduced number of Iba1 + cells when compared to the vehicle-treated group (Fig. 3C). A similar pattern was observed in CD4 + T cells, for which a significant reduction was observed after MAT treatment (Fig. 3D). These results indicate that MAT has a potent therapeutic effect in optic nerve inflammation.

MAT treatment decreases optic nerve demyelination
Once inflammation occurs, demyelination of the optic nerve begins and can be detected by LFB staining of myelin (12, 21). To assess demyelination of the optic nerve after MAT treatment, LFB staining was performed in all rats. Optic nerves of both vehicle-and MAT-treated rats displayed significantly reduced myelin staining compared with that of naïve animals, and demyelination was markedly decreased after MAT treatment compared to vehicle-treated rats (Fig. 4A, B). Thus, MAT treatment can effectively mitigate demyelination in the optic nerves of diseased rats.
3.4. MAT treatment reduces axonal loss in the optic nerve Neurofilaments (NFs), a major component of the neuronal cytoskeleton, are believed to function primarily to provide structural support for the axon and to regulate axon diameter (22,23). Optic nerve sections of all rats were therefore stained with NF antibody to test whether MAT can protect axons from inflammation-induced damage (Fig. 5A). Our results showed that NF expression was significantly decreased in optic nerves of vehicle-treated rats compared with naïve ones, while this expression was greatly profoundly restored after MAT treatment (Fig. 5B).

MAT treatment reduces RGC apoptosis
To determine whether MAT has an effect on decreasing RGC apoptosis and protecting RGC survival in ON, retinas were harvested from all rats and double staining was performed using anti-Brn3a (for RGCs) and TUNEL (for apoptosis) antibodies (Fig. 6A). Compared to naïve rats, the number of Brn3a+TUNEL+ in immunized rats was increased, while there was a significant decrease after MAT treatment compared with vehicle-treated rats (Fig. 6B). These results indicated that MAT treatment can reduce RGC apoptosis, and thus promote their survival.

Discussion
ON is characterized by inflammatory demyelination and axonal injury in the optic nerve, leading to RGC loss and visual dysfunction (24). ON commonly occurs in MS patients and in its animal model, EAE, as well (8). Previous studies have described the histopathological aspects of ON, but neuronal loss in animal models of experimental ON has been less well studied. It has been found that ON is not only an inflammatory condition, but also involves significant neurodegeneration (25); however, few therapies are known to be effective for RCG protection, and neuronal loss in animal models of experimental ON has not been well addressed. The goal of this study is to explore the potential antiinflammatory and neuroprotective activities of MAT in ON of EAE rats. We have in previous studies shown that treatment with MAT could prevent EAE (26,27); however, whether this natural alkaloid can protect neurons in ON is still unknown. Here we have for the first time provided evidence that MAT treatment resulted in clinical improvement in ON during EAE, as indicated by reduced inflammation and demyelination in the optic nerve [28]. The upregulated expression of neurofilaments and reduced RGC apoptosis after MAT treatment suggest that MAT has neuroprotective properties as well.
It has been shown that inflammatory responses play an important role in the development of ON (28), and optic nerve demyelination and infiltration have also been found to correlate with the severity of clinical disease in EAE mice (29). Among inflammatory cells, activated macrophages and microglia are the major cell types in ON that are closely associated with demyelination, axonal damage and loss of visual function (30). Indeed, when an inflammatory event occurs, such as autoimmunity, neural injuries or ischemia, microglia rapidly become activated and begin migrating to the event site while releasing pro-inflammatory substances such as TNF-α and interleukins that lead to tissue damage (31). Significantly more microglia have also been found in retinas of ON, which could be a direct response to RGC degeneration (32). T cells, by secreting proinflammatory cytokines, play a major role in the inflammatory demyelination of the optic nerve (33)(34)(35). Our data show increased numbers of CD4 + T cells and Iba1 + microglia/macrophages in the ON rats, which were significantly reduced after MAT treatment. The observation in the present study on the anti-inflammatory effects of MAT in experimental ON is consistent with findings in a variety of other inflammatory diseases and animal models. MAT possesses significant anti-hepatitis, immunosuppressive, anti-tumor, and anti-hepatic fibrosis capacities (36). Previous research has shown that MAT can inhibit immune activities of T cells, B cells and macrophages, at relatively low doses, and it is known to have partially suppressed development of EAE (18). In addition, MAT therapy significantly suppressed the production of proinflammatory cytokines, such as IFN-γ, TNF-α and IL-17, and blocked the migration of peripheral immune cells into the CNS (19), suggesting that it may be beneficial in ON.
NFs, which are synthesized in the neuron body and then transported into the axons, play a key role in the axonal cytoskeleton (37). In the present study, a decrease in the content of NFs was observed in the vehicle-treated EAE rats, while MAT treatment effectively restored its levels to baseline. These results suggest that an effective treatment can preserve this axon-associated protein from inflammation-induced damage. Consistent with our observations, phosphorylated neurofilament heavy chain was found increased in serum in an ON model of MOG-specific TCR transgenic mice, indicating the NFs were released into the bloodstream from damage to optic nerve axons (38). Indeed, a 3-fold reduction in NF levels has been revealed in the pooled optic nerve samples from both eyes of these ON mice, which is consistent with reduced visual function and optic nerve atrophy visualized by MRI (39). Similarly, a reduced level of NF expression was observed in spinal cord of untreated EAE mice, while this level was significantly increased after treatment, accompanied by improved clinical score of disease (40). On the other hand, demyelination of axons in the optic nerve results in apoptosis of RGCs, which is the major cause of vision loss in ON (8, 12, 41), and inhibition of proinflammatory signaling resulted in a nearly complete prevention of axonal demyelination, as well as a drastic attenuation of RGC death in ON (42). Consistent with these observations, we detected a large number of apoptotic RGCs in untreated rats, and the number was significantly reduced upon MAT treatment. These results, together with enhanced expression of NFs, suggest that MAT treatment reduces axonal loss, and then promotes RGC survival during experimental ON.
In summary, our study demonstrates that MAT effectively suppresses RGCs apoptosis in experimental ON, with suppression of inflammatory demyelination and axonal loss in optic nerves. This neuroprotective effect may be due to its anti-inflammatory effects and likely other mechanisms, e.g., a direct neuroprotective capacity. Taken as a whole, our study shows that MAT is a promising potential therapy that warrants further investigation for use in ON.

Consent to publish Not Applicable
Availability of data and materials The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that that they have no conflicts of interest.