Neuroprotective Effects of Fingolimod in a Cellular Model of Optic Neuritis

Visual dysfunction resulting from optic neuritis (ON) is one of the most common clinical manifestations of multiple sclerosis (MS), characterized by loss of retinal ganglion cells, thinning of the nerve fiber layer, and inflammation to the optic nerve. Current treatments available for ON or MS are only partially effective, specifically target the inflammatory phase, and have limited effects on long-term disability. Fingolimod (FTY) is an FDA-approved immunomodulatory agent for MS therapy. The objective of the current study was to evaluate the neuroprotective properties of FTY in the cellular model of ON-associated neuronal damage. R28 retinal neuronal cell damage was induced through treatment with tumor necrosis factor-α (TNFα). In our cell viability analysis, FTY treatment showed significantly reduced TNFα-induced neuronal death. Treatment with FTY attenuated the TNFα-induced changes in cell survival and cell stress signaling molecules. Furthermore, immunofluorescence studies performed using various markers indicated that FTY treatment protects the R28 cells against the TNFα-induced neurodegenerative changes by suppressing reactive oxygen species generation and promoting the expression of neuronal markers. In conclusion, our study suggests neuroprotective effects of FTY in an in vitro model of optic neuritis.


Introduction
Multiple sclerosis (MS) is an autoimmune disease of the central nervous system (CNS) prevalent in about 400,000 people in the US and 2.1 million people worldwide [1][2][3][4]. Approximately 20% of MS patients present with vision deficits associated with optic neuritis (ON) [5,6], and neurodegeneration characterized by loss of retinal ganglion cells, thinning of the nerve fiber layer, and axonal damage [7,8]. Parameters of visual function are utilized as necessary outcome measures in MS studies [9]. Although the current MS therapies target the inflammatory pathology, effects on the long-term neurodegenerative phases of the disease have not been shown. A treatment that effectively targets both aspects of MS would likely achieve preferred status as a disease-modifying agent.
Fingolimod (FTY720 or FTY), a sphingosine analog that functions as a potent immunosuppressive agent in the CNS, is approved for MS therapy, especially for highly active disease [10,11]. Once phosphorylated into its active form by the sphingosine-1-kinase, it acts as an agonist on sphingosine-1-phosphate (S1P) receptors [12][13][14][15] and induces internalization of S1P receptors after binding [16]. Pharmacologically FTY is known as an immunomodulatory drug. FTY exerts its immunosuppressive effect by lymphocyte sequestration and thus reduces the numbers of T and B cells in circulation [17,18]. FTY has been shown to exhibit neuroprotective properties in experimental models of Alzheimer's, stroke, Dose-response experiments were conducted to standardize the in vitro treatment with TNFα (recombinant rat tumor necrosis factor-α) (Cat # 510-RT, R&D Systems, Minneapolis, MN, USA) to induce neuronal injury in R28 cells. On day 0, cells were differentiated on 6-well culture plates (24 h), as described above. Treatments with TNFα at doses of 5, 10, 25, 50 ng/mL were initiated on day 1, followed by a 24 h incubation period. Cell viability with various doses of TNFα was compared against the control group with no treatment that depicted average growth and differentiation.

Treatments with Fingolimod
Once the effective dose of TNFα was established, experiments were conducted to identify an appropriate treatment concentration of FTY (Cat # 11975, Cayman Chemicals, Ann Arbor, MI, USA). Cells were pre-treated with FTY at concentrations of 2.5, 5, 10, 25, 50, and 100 nM for a 1 h incubation period before the TNFα treatment. Cell viability differences were compared among control (no treatment), TNFα-treated group, and TNFα co-incubated with varying FTY concentrations. FTY treatment alone at higher doses of 100, 200, and 500 nM were performed to test its cytotoxicity.

Cell Viability
The degree of viability of R28 cells post-treatment with TNFα and/or FTY was determined using the Trypan blue method [31]. Cells were seeded on six well plates at a density of 0.5 × 10 6 per well. Following treatments, they were trypsinized and collected in labeled tubes respective to their grouping. Equal volumes of a sample of cell suspension and trypan blue dye were thoroughly mixed using a micropipette, from which 10 µL was injected into a cell counting chamber (Cat # 02-671-55A, Fischer Scientific, Waltham, MA, USA) for manual counting. Trypan blue dye stains dead cells blue, and the number of viable cells in all four 16-squared tiles of the chamber were counted. This was repeated in triplicates for each cell suspension sample, and cell viability was plotted as the percentage with respect to 100% control. All graphs are represented as Mean ± SEM.

Western Blot Analysis
R28 cells seeded at a density of differentiated on six well plates and treated with TNFα and/or FTY as described earlier, resulting in four groups: Control (no treatment), TNFα, TNFα + FTY, and Control + FTY groups. Cells were homogenized, and the lysate was collected in RIPA buffer (Cat # 20-188, EMD Millipore, Burlington, MA, USA) containing protease (Cat # 78430, Fischer Scientific, Waltham, MA, USA) and phosphatase inhibitors (Cat # 78428, Fischer Scientific, Waltham, MA, USA). Protein estimation was performed using the Bradford's protein assay kit (Cat # 5000201, Bio-Rad Laboratories, Hercules, CA, USA). Samples with an equal amount of protein were prepared by using 4X Laemmli buffer (Cat # 161-0747, Bio-Rad Laboratories, Hercules, CA, USA) containing β-mercaptoethanol (Cat # O3446I-100, Fischer Scientific, Waltham, MA, USA). Samples were separated on SDS-PAGE and transferred to nitrocellulose membranes (Cat # 1620112, Bio-Rad Laboratories, Hercules, CA, USA). Membranes were blocked in 5% milk (Cat # 1706404, Bio-Rad Laboratories, Hercules, CA, USA) in tris-buffered saline with tween-20 (TBS-T) and incubated with respective primary antibodies (Table A1) overnight at 4 • C. Membranes were washed with 1× TBS-T and incubated in appropriate secondary antibodies. Signals were detected using enhanced chemiluminescence (ECL) (Cat # 32106, Fischer Scientific, Waltham, MA, USA). NIH Image J software was utilized to conduct densitometric analysis, and the intensity measurements were normalized to loading control. Experiments were repeated for a minimum of three times.

Immunofluorescence Staining
Cells seeded at a density of 15,000 to 20,000 cells per well on 8 well glass chamber slides (Cat # 154941, Fischer Scientific, Waltham, MA, USA) were treated according to the study design described previously. Following the 24 h incubation period, the culture media was removed, cells were washed with 1× PBS and fixed with 2% paraformaldehyde for 10 min. This was followed by a wash with PBS, and the chamber slides were stored in humidified containers at 4 ºC. Slides were brought to room temperature and washed with PBS before initiating the staining protocol. Permeabilization was achieved using 0.1% Triton X-100 in PBS for 5 min, followed by a PBS wash, and blocking with 10% donkey serum at room temperature for 1 h. Cells were washed and incubated with respective primary antibodies (Table A1) overnight. The next day, the cells were incubated with appropriate secondary antibodies for 2 h. The Slides were washed, and the chambers were separated from the glass slide. Cells were covered with a coverslip using a mounting medium containing DAPI and stored in 4 • C. Images were taken using a confocal microscope (LSM 780; Carl Zeiss, Thornwood, NY, USA) available at the Augusta University imaging core facility. For the Tuj1 and NSE quantitative analysis, similar thresholds were set for all the images. Three to five regions of interest (ROI) were randomly selected per chamber slide images, and the fluorescent intensity of immunoreactivity was measured (Integrated Density) using NIH Image J software. Along with the average fluorescence intensities of a given ROI, the average fluorescence intensity of areas without staining (background) was measured as well. The Corrected Fluorescence was calculated as Integrated Density-(Area of selected Cells 2021, 10, 2938 4 of 16 cells X Mean fluorescence of background readings). The values were then normalized relative to percentage of control group. Experiments were repeated a minimum of three times and details are provided under figure legends.

Cellular ROS Formation Using DCF Assay
CM-H2DCFDA (General Oxidative Stress Indicator, Thermofisher Scientific, cat# C6827) was used to measure reactive oxygen species (ROS) formation in TNFα treated cells and any changes in response to FTY treatment, as per the manufacturer's instructions. Briefly, cells were washed with ice-cold PBS (three times), and then incubated with 10 µm CM-H2DCFDA (working solution) at 37 • C for 30 min under dark. Washed with ice-cold PBS (three times), cover-slipped using the mounting medium with DAPI stain (Vector laboratories), and the images were immediately taken by confocal microscope (LSM 780; Carl Zeiss, Thornwood, NY, USA). NIH Image J software was used for the analysis of fluorescent intensity. For single-cell quantification, single-cell was delineated and sampled at 40× from a random start point. Only cells with precise neuronal shape and specific nuclear staining with DAPI were analyzed. Three to five chamber slides per treatment were examined and the experiments were repeated three times.

Mitochondrial ROS Measurement
The production of superoxide by mitochondria in response to TNFα treatment and the impact of FTY on the mitochondrial ROS generation was measured using the MitoSOX™ Red reagent (Thermo Fisher Scientific, cat# M36008), following the manufacturer's instructions. Briefly, cells on chamber slides were washed with ice-cold PBS (three times), incubated with 5 µM MitoSOX™ reagent working solution, and incubated for 10 min at 37 • C, protected from light. Chamber slides were then carefully washed with ice-cold PBS (three times), cover slipped, and images were immediately taken by confocal microscope (LSM 780; Carl Zeiss, Thornwood, NY, USA). NIH Image J software was used for the analysis of fluorescent intensity. Three to five chamber slides per treatment were examined and the experiments were repeated three times.

H 2 O 2 Treatment and LDH Assay
Oxidative stress was induced on differentiated R28 cells using H 2 O 2 (SigMA, USA) following the method of Song et al. [32] with some minor modifications. Briefly, R28 cells were treated with multiple concentrations of H 2 O 2 (0.0, 0.2, 0.4, 0.6, 0.8, 1.0, and 1.5 mM) for 24 h and cellular cytotoxicity was measured. In order to determine the effect of FTY on H 2 O 2 -induced oxidative stress, cells were pre-treated with 25 nM FTY (as described previously) and changes in cytotoxicity was measured.
Lactate dehydrogenase (LDH) assay was used to determine cellular cytotoxicity. LDH released into the culture media from damaged cells was measured following the manufacturer's instructions (CytoTox 96 non-radioactive cytotoxicity assay kit; Promega Corporation, Madison, WI, USA). The level of LDH release was normalized to the total LDH content following cell lysis in a medium. The absorbance was determined at 490 nm using a Multimode Microplate Reader (Berthold Technologies, Bad Wildbad, Germany). LDH release was expressed as a percentage of the maximum LDH released after cell lysis.

Statistical Analysis
All statistical analyses were performed with GraphPad Prism 7 (GraphPad Software Inc., La Jolla, CA, USA). Student t-test or Two-way ANOVA followed by Tukey's multiple comparisons test was employed to analyze the groups. A p value less than 0.05 was considered as statistically significant. Results are presented as Mean ± SEM.

Fingolimod Treatment Reduces TNFα-Induced Neuronal Injury
We found that TNFα treatment resulted in a significant reduction (p < 0.05) in cell viability at doses of 10 ng/mL (44.6 ± 24.8%), 25 ng/mL (39.8 ± 8.7%), and 50 ng/mL (60.2 ± 13.0%) compared to the untreated control ( Figure 1A). Our findings suggested that TNFα at 10 ng/mL desirably reduced the percentage of viable cells by nearly half that of the control group. In the next step, cells were pre-treated with fingolimod at concentrations of 0, 2.5, 5, 10, 25, 50, and 100 nM for 1 h prior to TNFα induction (10 ng/mL). Our results showed that FTY concentrations at 25 nM (79.7 ± 17.7%) and 50 nM (71.0 ± 32.3%) significantly prevented the TNFα-induced injury (p < 0.001) ( Figure 2B). Experiments with high-dose FTY treatment alone resulted in cytotoxicity at doses above 100 nM. Doses of 100, 200, and 500 nM showed a decreasing viable cell count of 14.5 × 10 4 , 3.5 × 10 4 , and 2 × 10 4 , respectively, compared to an average viable cell count of 47.5 × 10 4 in control cells without FTY treatment (not shown). Based on our findings, 25 nM was chosen as the dose of FTY to be used in further studies ( Figure 1C).

Fingolimod Treatment Reduces TNFα-Induced Neuronal Injury
We found that TNFα treatment resulted in a significant reduction (p < 0.05) in cell viability at doses of 10 ng/mL (44.6 ± 24.8%), 25 ng/mL (39.8 ± 8.7%), and 50 ng/mL (60.2 ± 13.0%) compared to the untreated control ( Figure 1A). Our findings suggested that TNFα at 10 ng/mL desirably reduced the percentage of viable cells by nearly half that of the control group. In the next step, cells were pre-treated with fingolimod at concentrations of 0, 2.5, 5, 10, 25, 50, and 100 nM for 1 h prior to TNFα induction (10 ng/mL). Our results showed that FTY concentrations at 25 nM (79.7 ± 17.7%) and 50 nM (71.0 ± 32.3%) significantly prevented the TNFα-induced injury (p < 0.001) ( Figure 2B). Experiments with highdose FTY treatment alone resulted in cytotoxicity at doses above 100 nM. Doses of 100, 200, and 500 nM showed a decreasing viable cell count of 14.5 × 10 4 , 3.5 × 10 4 , and 2 × 10 4 , respectively, compared to an average viable cell count of 47.5 × 10 4 in control cells without FTY treatment (not shown). Based on our findings, 25 nM was chosen as the dose of FTY to be used in further studies ( Figure 1C).

Fingolimod Attenuates Cellular Stress and Survival Signaling
Changes in phosphorylated p38 MAP kinase expression were assessed to characterize cellular stress by Western blotting. Figure 2A shows that TNFα augmented the level of p-P38 MAPK, and this increase was markedly prevented in the presence of FTY. Moreover, we found that FTY treatment alone did not affect levels of p-P38 MAPK. Our quantification data demonstrated that in the presence of FTY, levels of p-P38/total-P38 were significantly reduced (p < 0.05) versus the TNFα group ( Figure 2B). Changes in phosphorylated Akt levels were tested to assess cell survival using Western analysis. Figure 2C shows no changes in the expression of p-AKT with TNFα induction. Consistently, our quantification data also showed no significant decrease in levels of p-Akt/t-Akt in the presence of TNFα versus control ( Figure 2D).

Fingolimod Attenuates Cellular Stress and Survival Signaling
Changes in phosphorylated p38 MAP kinase expression were assessed to characterize cellular stress by Western blotting. Figure 2A shows that TNFα augmented the level of p-P38 MAPK, and this increase was markedly prevented in the presence of FTY. Moreover, we found that FTY treatment alone did not affect levels of p-P38 MAPK. Our quantification data demonstrated that in the presence of FTY, levels of p-P38/total-P38 were significantly reduced (p < 0.05) versus the TNFα group ( Figure 2B). Changes in phosphorylated Akt levels were tested to assess cell survival using Western analysis. Figure 2C shows no changes in the expression of p-AKT with TNFα induction. Consistently, our quantification data also showed no significant decrease in levels of p-Akt/t-Akt in the presence of TNFα versus control ( Figure 2D).

Effect of Fingolimod on Neuronal Cell Death
To evaluate apoptotic changes, Western blot analyses using apoptotic marker cleaved caspase-3, and anti-apoptotic marker Bcl-xL were performed. An upregulated expression on cleaved caspase-3 along with a decreased level of Bcl-xL in the presence of TNFα was observed. In the co-treatment group with TNFα and FTY, we observed a reversal in these changes ( Figure 3). Quantification data showed a significant increase in the levels of cleaved caspase-3 (p < 0.01) versus control, while FTY treatment significantly reversed this effect (p < 0.05) versus the TNFα group ( Figure 3B). TNFα caused a significant decrease in levels of Bcl-xL (p < 0.05) compared to the control group. However, the difference observed in response to FTY co-treatment was not statistically significant compared to the TNFα group ( Figure 3D).

Effect of Fingolimod on Neuronal Cell Death
To evaluate apoptotic changes, Western blot analyses using apoptotic marker cleaved caspase-3, and anti-apoptotic marker Bcl-xL were performed. An upregulated expression on cleaved caspase-3 along with a decreased level of Bcl-xL in the presence of TNFα was observed. In the co-treatment group with TNFα and FTY, we observed a reversal in these changes ( Figure 3). Quantification data showed a significant increase in the levels of cleaved caspase-3 (p < 0.01) versus control, while FTY treatment significantly reversed this effect (p < 0.05) versus the TNFα group ( Figure 3B). TNFα caused a significant decrease in levels of Bcl-xL (p < 0.05) compared to the control group. However, the difference observed in response to FTY co-treatment was not statistically significant compared to the TNFα group ( Figure 3D).

Effect of Fingolimod on TNFα-Induced Neuronal Damage
Immunofluorescence studies were conducted to study the neurodegenerative changes observed in R28 cells in response to the different treatments. β-tubulin III, also called Tuj1 contributes to microtubule formation in neuronal cell bodies and axons and plays important roles in axonal transport and cell differentiation. It is a useful marker for the detection of injury-related alterations [33]. Neuron specific enolase (NSE) is widely used and accepted as a neuronal marker, and is expressed by mature neurons and cells of neuronal origin [34]. In the present study, immunostaining with Tuj1 and NSE revealed the degenerative changes induced by TNFα, which was attenuated with FTY treatment ( Figure 4A,C). Tuj1 expression was downregulated in TNFα-treated cells; however, we observed that FTY was able to protect the cells against neurofilament damage ( Figure 4A lower panels). Consistently, the NSE marker was found to be reduced in the presence of TNFα, and the reduction was prevented by FTY treatment (Figure 4B,D).

Effect of Fingolimod on TNFα-Induced Neuronal Damage
Immunofluorescence studies were conducted to study the neurodegenerative changes observed in R28 cells in response to the different treatments. β-tubulin III, also called Tuj1 contributes to microtubule formation in neuronal cell bodies and axons and plays important roles in axonal transport and cell differentiation. It is a useful marker for the detection of injury-related alterations [33]. Neuron specific enolase (NSE) is widely used and accepted as a neuronal marker, and is expressed by mature neurons and cells of neuronal origin [34]. In the present study, immunostaining with Tuj1 and NSE revealed the degenerative changes induced by TNFα, which was attenuated with FTY treatment (Figure 4A,C). Tuj1 expression was downregulated in TNFα-treated cells; however, we observed that FTY was able to protect the cells against neurofilament damage ( Figure 4A lower panels). Consistently, the NSE marker was found to be reduced in the presence of TNFα, and the reduction was prevented by FTY treatment (Figure 4B,D).

Effect of Fingolimod on ROS Formation
Changes in ROS formation were studied using DCF assay. As shown in Figure 5A, ROS levels were observed to be markedly elevated in the neuronal cells in response to TNFα treatment. However, treatment with FTY downregulated the TNFα induced ROS generation. Our quantification studies demonstrate that the ROS levels are significantly higher in the TNFα group compared to control and are significantly reduced in response to FTY co-treatment ( Figure 5B).

Effect of Fingolimod on ROS Formation
Changes in ROS formation were studied using DCF assay. As shown in Figure  ROS levels were observed to be markedly elevated in the neuronal cells in response TNFα treatment. However, treatment with FTY downregulated the TNFα induced R generation. Our quantification studies demonstrate that the ROS levels are significan higher in the TNFα group compared to control and are significantly reduced in respo to FTY co-treatment ( Figure 5B).

Effect of Fingolimod on Mitochondrial Dynamics and ROS Formation
In the present study, we investigated the effect of FTY treatment on the changes proteins related to mitochondrial dynamics, including DRP-1 (dynamin related prote 1), Mitofusin 2 and OPA-1 (optic atrophy 1). As illustrated in Figure 6A,D, expression Mitofusin 2 was significantly reduced in TNFα treated R28 cells, while FTY treatment n malized the level of Mitofusin 2 in TNFα treated cells, similar to control levels. An incre in the level of p-DRP1 was observed in TNFα treated group, while treatment with F reversed this effect. However, these changes were not statistically significant. No mark differences were seen in the other mitochondrial proteins studied. Further, we inve gated the impact of FTY treatment on mitochondrial ROS formation using MitoSox as ( Figure 6G,H). TNFα treatment resulted in the generation of mitochondrial ROS as e denced by elevated fluorescence indicator. FTY treatment markedly reduced the leve mitochondrial ROS formed in response to TNFα treatment. Quantification results dem strate that the mitochondrial ROS level is significantly upregulated in TNFα treated gro and the treatment with FTY significantly reduced the effect ( Figure 6H).

Fingolimod Attenuates H2O2-Induced Cellular Damage and Stress Signaling in R28 Cells
To further assess the neuroprotective effect of FTY, experiments were performed ing H2O2, another cellular stressor. Our results show that H2O2 induces cytotoxicity in ferentiated R28 cells in a dose-dependent manner ( Figure 7A). Treatment with FTY sig icantly reduced the cytotoxicity induced by TNFα at the various concentrations stud ( Figure 7B). Our results indicate that H2O2 treatments significantly elevated p-P38 lev at all the concentrations studied, while FTY treatment significantly reduced the effec two different contentions of H2O2. FTY treatment did not offer protection at a higher c centration, 1mM of H2O2 ( Figure 7C).

Effect of Fingolimod on Mitochondrial Dynamics and ROS Formation
In the present study, we investigated the effect of FTY treatment on the changes in proteins related to mitochondrial dynamics, including DRP-1 (dynamin related protein-1), Mitofusin 2 and OPA-1 (optic atrophy 1). As illustrated in Figure 6A,D, expression of Mitofusin 2 was significantly reduced in TNFα treated R28 cells, while FTY treatment normalized the level of Mitofusin 2 in TNFα treated cells, similar to control levels. An increase in the level of p-DRP1 was observed in TNFα treated group, while treatment with FTY reversed this effect. However, these changes were not statistically significant. No marked differences were seen in the other mitochondrial proteins studied. Further, we investigated the impact of FTY treatment on mitochondrial ROS formation using MitoSox assay ( Figure 6G,H). TNFα treatment resulted in the generation of mitochondrial ROS as evidenced by elevated fluorescence indicator. FTY treatment markedly reduced the level of mitochondrial ROS formed in response to TNFα treatment. Quantification results demonstrate that the mitochondrial ROS level is significantly upregulated in TNFα treated group and the treatment with FTY significantly reduced the effect ( Figure 6H).

Fingolimod Attenuates H 2 O 2 -Induced Cellular Damage and Stress Signaling in R28 Cells
To further assess the neuroprotective effect of FTY, experiments were performed using H 2 O 2 , another cellular stressor. Our results show that H 2 O 2 induces cytotoxicity in differentiated R28 cells in a dose-dependent manner ( Figure 7A). Treatment with FTY significantly reduced the cytotoxicity induced by TNFα at the various concentrations studied ( Figure 7B). Our results indicate that H 2 O 2 treatments significantly elevated p-P38 levels at all the concentrations studied, while FTY treatment significantly reduced the effect at two different contentions of H 2 O 2. FTY treatment did not offer protection at a higher concentration, 1mM of H 2 O 2 ( Figure 7C).

Discussion
The present study was conducted to assess the potential neuroprotective action of FTY in an in vitro model of optic neuritis. Lack of effective treatment strategies to reduce neurodegeneration continues to be a major problem in the field of MS research. It is vital to understand the underlying mechanisms of MS-induced neuronal damage and dysfunction. Even though MS research on the pathophysiology and associated molecular mechanisms have evolved over decades of research, including our laboratory [35,36], the field lacks reliable in vitro models to study neurodegeneration. Synthetic molecules such as trimethyltin [37], oxaliplatin [38], and cuprizone [39], although successful in creating a neurodegenerative environment, do not accurately represent the neuroinflammatory changes observed in an MS brain. In MS, inflammatory leukocytes are believed to infiltrate the CNS to mediate demyelination and neuronal degeneration via cytokines upon activation of T lymphocytes and antigen-presenting cells (APCs) [40]. tumor necrosis factor-α (TNFα) is one of the primary cytokines that are present in elevated levels in active MS lesions, serum, and cerebrospinal fluid of MS patients [41]. Studies conducted on BV-2 microglial cell lines [42] and primary mixed neuronal and glial cultures [43] show the effect of TNFα-induced damage and apoptosis. Hence, utilizing the R28 neuro-retinal cells, we standardized an in vitro experimental model of neurodegeneration to assess the impact of fingolimod on TNFα-induced neuronal damage. Using a set of functional and biochemical analyses, our study demonstrates the neuroprotective properties of fingolimod in MS-associated optic neuritis in vitro.
We first demonstrated the dose-dependent effects of TNFα on neuronal cell viability in R28 neuro-retinal cells in vitro and identified the optimal dose of TNFα for further molecular characterization. The R28 cells are immortalized, rat retinal origin, heterogenous, precursor cells with differentiation potential [26]. According to the published methods, R28 cells were differentiated to neuronal phenotype with the addition of a modified form of cAMP and laminin and grown using DMEM [44][45][46]. Studies by Seigel et al. utilizing R28 cells demonstrate the expression of neuron-specific markers such as MAP2, Syntaxin, NSE, and Nestin, along with neurotransmission receptors [26,28]. Our study characterized the expression of neurofilament marker Tuj1, along with NSE in these cells, thus demonstrating the reliability of this model in evaluating the impact of FTY on the neuronal injury. In MS, one of the major clinical presentations observed in patients is optic neuritis [47]. Studies have shown that approximately 20% of patients present with inflammation of the optic nerve as their first symptom of MS [5,6]. Another study conducted by the North American Research Committee on multiple sclerosis (NARCOMS) showed that of the 9107 patients participating in the study, 60% reported signs of vision impairment, and 14% of these depicted moderate/severe/very severe impairment of vision [9]. Based on the available research, visual dysfunction is a common component of MS disease progression and an important determinant of quality of life. Current MS therapies function by suppressing the inflammatory pathways and have unknown impact on the long-term neuronal damage, causing a major knowledge gap and emphasizes the need to identify a neuroprotective therapeutic agent. Therefore, our study focused on assessing the neuroprotective effect of FTY in an in vitro model of MS-induced optic neuritis. Fingolimod, a sphingosine-1-phosphate (S1P) receptor modulator, has previously been shown to prevent neurodegenerative mechanisms targeting an inflammatory CNS state in vitro, in vivo, and clinical settings, as detailed below. Studies on Parkinson's disease models have shown a positive impact with FTY [19][20][21]. Mechanistically, it was found that the protective effects of FTY in Parkinson's disease were correlated with the activation of survival pathway mediated by Akt/ERK1/2 and increased expression of a neuron-specific brain-derived neurotrophic factor (BDNF). In a model of Alzheimer's, FTY was able to reverse the effect of damage by modulating the levels of different markers such as the Glial fibrillary acidic protein (astrogliosis marker), taurine (anti-inflammatory marker), and neuronal markers such as the N-acetyl aspartate and glutamate [23]. A meta-analysis was conducted by Liu et al. which included nine studies that focused on quantification of infarct volume and neurological deficit scoring in a transient middle cerebral artery occlusion model of ischemic stroke challenged with FTY [24]. The study concluded that FTY could be a possible candidate for stroke due to its protective effects on neurological deficit and infarct volume in eight of the nine included studies. Promising outcomes with FTY represent the need for further investigation to confirm the theories on its action as a neuroprotective agent. Utilizing the EAE (Experimental Autoimmune Encephalitis) model of MS, a recent study by Yang et al. investigated whether FTY is beneficial to the visual system [25]. Their results showed that FTY treatment offered neuroprotective and anti-inflammatory effects on the retina and optic nerve. FTY treatment alleviated EAE-induced gliosis, inflammation and reduced the apoptosis of RGCs and oligodendrocytes.
In response to FTY treatment, our studies found a reduction in the phosphorylation of p38 MAP kinase (a cellular stress signaling pathway) in TNFα-treated R28 cells. However, fingolimod treatment did not induce any changes in the levels of phosphorylated Akt, indicating its effect on cell survival is independent of Akt, with the conditions studied. TNFα-induced cell death was confirmed by the upregulation of cleaved caspase-3 (a cell death marker) expression along with reduced levels of Bcl-xL (an anti-apoptotic protein). These changes were reversed in response to FTY treatment, supporting its neuroprotective and anti-apoptotic function. FTY treatment also protected the retinal neurons against the TNFα-induced neuronal damage determined by the expression of neuronal markers.
Oxidative stress plays a critical part in the pathogenesis of various neurodegenerative diseases and neuroinflammation. There exists an increasing amount of data indicating that oxidative stress plays a major role in the pathogenesis of MS and optic neuritis [48]. Results from preclinical studies show that suppression of oxidative stress is a promising strategy for optic neuritis [49][50][51][52][53]. In the present study, we show that FTY mediated neuroprotection of R28 cells involves the regulation of ROS formation. A recent study performed on serum samples from patients with relapsing-remitting MS and healthy controls demonstrated that TFY treatment increased total antioxidant capacity [54]. Furthermore, FTY treatment reduced oxidative stress in experimental models of cardiomyopathy [55], multiple system atrophy [56], autism [57], and vitamin K-induced neurotoxicity [20].
Oxidative damages along with mitochondrial dysfunction are common characteristics of neurodegenerative diseases. Mitochondrial dysfunction is increasingly recognized as a major mechanism of MS associated pathologies [58][59][60]. The present study investigated the impact of FTY treatment on the changes in protein levels related to mitochondrial dynamics. Alterations in mitochondrial dynamics affect mitochondrial size and shape and impact mitochondrial metabolism and cell death. These events are controlled by mitochondrial dynamin-related GTPases, including mitofusin-1, mitofusin-2, OPA1, and DRP1. Our results show that one possible mechanism by which FTY offers neuroprotection of R28 neuronal cells via regulation of mitochondrial fusion. Altered levels of mitochondrial proteins were reported in retinal neurons in models of diabetic retinopathy [61] and glaucoma [62]. Results from our study are consistent with studies on neuroprotective properties of FTY in other models where FTY improved mitochondrial stability and restored mitochondrial dynamics under oxidative stress conditions [20,[63][64][65]. Our present study did not investigate the changes in mitochondrial function.
Our study showed that FTY treatment reduces TNFα induced mitochondrial ROS formation in R28 cells. Other studies recently reported that FTY reduced mitochondrial dysfunction in a rat model of chronic cerebral hypoperfusion, reduced neuroinflammation and restored mitochondrial function in a model of Multiple System Atrophy [65], prevented mitochondrial dysfunction and protected neurons in prion protein-disease model [66]. Furthermore, we investigated the neuroprotective effect of FTY on R28 cells treated with H 2 O 2 , a cellular stressor known to induce oxidative stress. Our results reveal that FTY offers neuroprotection in response to oxidative stress-induced cellular damage. While FTY treatment rescued R28 cells at all concentrations of H 2 O 2 studied, FTY did not significantly reduce phospho-P38 levels at higher concentration of H 2 O 2 , suggesting the possibility of other signaling pathways involved. Another possibility is that FTY at a different dose/time could reduce the phospho-P38 levels at higher concentrations of H 2 O 2 . Overall, these results indicate that FTY mediated neuroprotection could be offered through multiple mechanisms, including P38MAPK signaling. However, further studies are needed to confirm this observation.
Evidence as a regulator of oxidative stress, along with its immunomodulatory function, offers significant therapeutic potential to FTY in neuroinflammatory diseases such as optic neuritis. Figure 8 depicts the possible mechanisms of FTY mediated neuroprotection in response to TNFα-induced damage. Further studies are needed to define whether the effect of FTY on ROS level is direct or indirect and to delineate the molecular mechanisms of FTY mediated neuroprotection.
Cells 2021, 10, 2938 12 o prevented mitochondrial dysfunction and protected neurons in prion protein-dise model [66]. Furthermore, we investigated the neuroprotective effect of FTY on R28 c treated with H2O2, a cellular stressor known to induce oxidative stress. Our results rev that FTY offers neuroprotection in response to oxidative stress-induced cellular dama While FTY treatment rescued R28 cells at all concentrations of H2O2 studied, FTY did significantly reduce phospho-P38 levels at higher concentration of H2O2, suggesting possibility of other signaling pathways involved. Another possibility is that FTY at a ferent dose/time could reduce the phospho-P38 levels at higher concentrations of H2 Overall, these results indicate that FTY mediated neuroprotection could be offe through multiple mechanisms, including P38MAPK signaling. However, further stud are needed to confirm this observation. Evidence as a regulator of oxidative stress, along with its immunomodulatory fu tion, offers significant therapeutic potential to FTY in neuroinflammatory diseases su as optic neuritis. Figure 8 depicts the possible mechanisms of FTY mediated neuropro tion in response to TNFα-induced damage. Further studies are needed to define whet the effect of FTY on ROS level is direct or indirect and to delineate the molecular mec nisms of FTY mediated neuroprotection. One limitation of our study is that it did not elucidate the role of cell survival pa ways that are directly associated with TNFα and FTY action. The concentration of F used in our study (25 nM) corresponds to 7.6875 ng/mL. However, as per the manuf turer's (Gilenya ® ) package insert, the concentration of active fingolimod phosphate adult MS patients is 1.35 ng/mL. Lower concentrations used in our study, did not o any neuroprotective effect. It is likely that the neurovascular unit could respond dif ently in response to other cytokines or injury mediators. This difference in human s temic concentration versus in vitro concentration in our models is an interesting aspec the study and suggests that repurposing may be needed for the neuroprotective action FTY in MS patients. Another limitation of the current study is that it is purely perform in a cellular model in vitro, which is not a true reflection of what happens in a complex vivo set up. Nevertheless, the study provides reasonable optimism on the potential th apeutic benefits of FTY to treat ON. Studies performed on EAE model by Yang el demonstrated neuroprotective and anti-inflammatory effects on the retina and op nerve, with no direct negative effects at the two different doses of FTY (0.3 and 1.0 mg/ utilized [25]. Results from our study complements the findings of Yang et al., wh stronger neuroprotective effects on the visual system. One limitation of our study is that it did not elucidate the role of cell survival pathways that are directly associated with TNFα and FTY action. The concentration of FTY used in our study (25 nM) corresponds to 7.6875 ng/mL. However, as per the manufacturer's (Gilenya ® ) package insert, the concentration of active fingolimod phosphate in adult MS patients is 1.35 ng/mL. Lower concentrations used in our study, did not offer any neuroprotective effect. It is likely that the neurovascular unit could respond differently in response to other cytokines or injury mediators. This difference in human systemic concentration versus in vitro concentration in our models is an interesting aspect of the study and suggests that repurposing may be needed for the neuroprotective action of FTY in MS patients. Another limitation of the current study is that it is purely performed in a cellular model in vitro, which is not a true reflection of what happens in a complex in vivo set up. Nevertheless, the study provides reasonable optimism on the potential therapeutic benefits of FTY to treat ON. Studies performed on EAE model by Yang el tal demonstrated neuroprotective and anti-inflammatory effects on the retina and optic nerve, with no direct negative effects at the two different doses of FTY (0.3 and 1.0 mg/kg) utilized [25]. Results from our study complements the findings of Yang et al., where stronger neuroprotective effects on the visual system.

Conclusions
Overall, our study investigated the potential neuroprotective effects of FTY in an in vitro model of neurodegeneration. The R28 neuro-retinal cells are characterized as a successful platform for evaluating neuronal damage in the presence of TNFα, and its suppression with FTY. Furthermore, our studies demonstrated the antioxidant properties of FTY, a possible mechanism of neuroprotection. However, further studies are required to confirm the results. Based on the cellular and molecular analysis, FTY demonstrated the potential to be investigated as a novel neuroprotective strategy in conditions like MS and associated pathologies. Funding: This study was supported in part by University of Georgia startup funds and the National Eye Institute to SPN (R01EY028569). This work has been accomplished using the resources and facilities at the Charlie Norwood VA Medical Center, Augusta, GA, and a core grant from the NIH/NEI to the Augusta University Vision Discovery Institute (P30EY031631). The funders had no role in study design, data collection, analysis, and decision to publish the data. The contents of the manuscript do not represent the views of the Department of Veteran Affairs or the United States.
Institutional Review Board Statement: Not applicable.

Informed Consent Statement: Not applicable.
Data Availability Statement: Not applicable.

Acknowledgments:
The authors are thankful to Chithra Palani for her assistance with the R28 cell culture studies.

Conflicts of Interest:
The authors declare no conflict of interest.