Apigenin Alleviates Autoimmune Uveitis by Inhibiting Microglia M1 Pro-Inflammatory Polarization

Abstract

Purpose: Apigenin is a natural small molecule compound widely present in various vegetables and fruits. Recently, Apigenin was reported to inhibit lipopolysaccharide (LPS)-simulated microglial proinflammatory activation. Considering the important role of microglia in retinal disorders, we wonder whether Apigenin could exert a therapeutic effect on experimental autoimmune uveitis (EAU) through reprogramming retinal microglia to a beneficial subtype.

Methods: EAU was induced in C57BL/6J mice by immunization with interphotoreceptor retinoid-binding protein (IRBP)_651–670, followed by intraperitoneal administration of Apigenin. Disease severity was assessed based on clinical and pathological scores. In vivo, Western blotting was used to quantify protein levels of classical inflammatory factors, microglial M1/M2 markers and the tight junction protein of the blood-retinal-barrier (BRB). Immunofluorescence was used to determine the Apigenin's efficacy on microglial phenotype. In vitro, Apigenin was added in LPS and IFN-γ stimulated human microglial cell line. Western blotting and Transwell assays were used to analyze the phenotype of microglia.

Results: In vivo, we found that Apigenin significantly reduced the clinical and pathological scores of EAU. The protein levels of inflammatory cytokines were significantly decreased in retina, and BRB disruption was ameliorated after Apigenin treatment. Meanwhile, Apigenin inhibited microglia M1 transition in EAU mice retina. In vitro functional studies showed that Apigenin decreased LPS and IFN-γ–induced microglial inflammatory factor production and M1-activation via the TLR4/MyD88 pathway.

Conclusions: Apigenin can ameliorate retinal inflammation in IRBP induced autoimmune uveitis through inhibiting microglia M1 pro-inflammatory polarization via TLR4/MyD88 pathway.

Autoimmune uveitis is a serious vision-threatening eye disease characterized by the infiltration of leukocytes in the retina and destruction of the light-sensitive photoreceptor cells. These infiltrating inflammatory cells then lead to the development of posterior uveitis or panuveitis.^1–3 It is an immune systemic disorder associated with a variety of peripheral and local immune cells.^4,5 The current clinical methods for the treatment of autoimmune uveitis are mainly the use of corticosteroids and immunomodulatory drugs, but long-term use would cause liver and kidney toxicity,⁶ so it is urgent to develop a drug with fewer side effects. 

Experimental autoimmune uveitis (EAU) is a widely used model of clinical uveitis that can be induced by immunization with interphotoreceptor retinoid-binding protein (IRBP).⁷ Microglia are the resident immune cells in the retina and central nervous system, derived from primitive macrophage during the embryonic period.⁸ They make irreplaceable contributions to the maintenance of normal homeostasis and immune surveillance of the retinal microenvironments.^9,10 However, they secrete various inflammatory mediators in the pathological state, resulting in severe tissue damage.¹¹ In the course of EAU, microglia transfer to the M1 phenotype, the classically activated phenotype that initiates inflammation and enhances retinal disturbance. Under the M1 subtype, the microglia undergo morphologic alterations, switching to ameboid shapes and migrate to the outer nuclear layer and subretinal regions.¹² However, another phenotype, M2, is a form that can promote repair and homeostasis.¹³ The inflammatory cytokines associated with the M1 microglia include IL-1β, IL-6, TNF-α, and inducible nitric oxide synthase (iNOS).^14,15 They can damage retinal photoreceptors, destroy the blood-retinal barrier (BRB), and allow peripheral lymphocytes to accumulate into the retina, leading to subsequent inflammatory cascade.^16,17 Considering the key role of microglia in autoimmune uveitis, promoting the transformation of microglia from the M1 to the M2 phenotype may be a potential therapeutic strategy. 

Apigenin, a flavonoid compound obtained from several fruits and vegetables, has been reported to exhibit antioxidant, anti-inflammatory activities in a variety of systematic disorders and several immune cells.^18–20 In high-throughput macrophage phenotype screening research, Apigenin was a potent drug capable of reprogramming resting macrophages to the M2 phenotype.²¹ Another study demonstrated that Apigenin could reduce murine BV2 microglia M1 activation after stimulation through the GSK3β/Nrf2 signaling pathway.²² Apigenin has been reported to be effective in several ocular diseases including AMD and retinal degeneration, demonstrating good biosafety. ^23,24 However, the therapeutic potential of Apigenin in autoimmune uveitis remains unclear. 

In our study, we used multiple indicators, including clinical and pathological scores, and expression levels of representative inflammatory factors, revealing the therapeutic role of Apigenin in EAU. Through building the IRBP-induced EAU model and using lipopolysaccharide (LPS) and IFN-γ activated human microglial cell line (HMC3) cells, we reported that Apigenin could inhibit microglial proinflammatory activation and alleviate intraocular inflammation in EAU through TLR4/MyD88 pathway. 

Material and Methods

Reagents

Human IRBP_651-670 (LAQGAYRTAV-DLESLASQLT) (purity > 98%) was synthesized by Sangon Biotech (Shanghai, China). Pertussis toxin was obtained from Sigma-Aldrich (St. Louis, MO, USA). A heat-killed Mycobacterium tuberculosis strain H37Ra was obtained from BD Biosciences (Franklin Lake, NJ, USA). DMSO, lipopolysaccharide, and complete Freund's adjuvant were purchased from Sigma-Aldrich. Recombinant human IFN-γ was purchased from PeproTech (Cranbury, NJ, USA). Apigenin (purity = 99.66%) was purchased from Topscience (Shanghai, China). 

Animal

Female C57BL/6J mice (six to eight weeks of age) were purchased from The Experimental Animal Center of Chongqing Medical University and kept under specific pathogen-free conditions at the Animal Care Service of Chongqing Medical University. All animal experiments conformed to the management of the Ethics Committee of the First Affiliated Hospital of Chongqing Medical University (approval number: 2019-296) and to the Association for Research in Vision and Ophthalmology Statement for the Use of Animals in Ophthalmic and Vision Research. 

Induction of EAU

Human IRBP_651–670 (5 mg) was dissolved in PBS (1 mL) and emulsified with complete Freund's adjuvant containing 5.0 mg/mL M. tuberculosis strain H37Ra (1:1, v/v) for one hour.^25–29 Mice received subcutaneous injection of IRBP (500 µg) and intraperitoneal injection of pertussis toxin (1 µg) for the EAU model. All mice were randomly divided into three groups: one EAU group received intraperitoneal administration of Apigenin (20 mg/kg body weight, from day 7 to day 13 after immunization once daily); the second EAU group received intraperitoneal administration of vehicle saline solution + 2% DMSO in the same volume (100 µL every mouse); and the nonimmunized mice were considered as the control group. From day 7, clinical scores of anterior segments were assessed daily according to the five independent criteria³⁰ in a blinded manner. At day 14, fundus images were taken, fundus clinical scores were assessed according to Caspi's criteria³¹ in a blinded manner, and then the mice were sacrificed for the following experiments. 

Hematoxylin and Eosin (H&E) Staining

At day 14, the eyeballs from three groups were harvested and fixed in 4% glutaraldehyde (Biosharp, Hefei, China) and embedded in paraffin. Tissue sections of 5 µm were cut and stained with H&E. Histological inflammation was graded on a scale between 0 and 4 in half-point increments according to Caspi's criteria³¹ in a blinded manner. 

Immunofluorescence Staining

The retinal paraffin sections and the retinal whole mounts were permeabilized with 3% Triton X-100 for 15 minutes and blocked with goat serum for two hours. Next, the retinal sections and flat mounts were incubated with anti-IBA1-rabbit IgG (1:1000; FUJIFILM Wako Chemicals, Japan), anti-IBA1-mouse IgG (1:100; Abcam, Cambridge, MA, USA), anti-CD74-rabbit IgG (1:100; Abcam), at 4°C overnight. After being washed with PBS three times, the retinal sections and flat mounts were incubated with Alexa-488 (green)–conjugated goat anti-rabbit IgG (1:500; Beyotime Institute of Biotechnology, Jiangsu, China) and Cy3 (red) conjugated goat anti-mouse IgG (1:500; Beyotime Institute of Biotechnology) at room temperature for one hour. For the retinal paraffin sections, nuclei were also stained with 4ʹ,6-diamidino-2-phenylindole (DAPI) (Beyotime Institute of Biotechnology). Images were taken under a microscope (Leica, Wetzlar, Germany). 

Cell Culture and Treatment

HMC3 was obtained from the American Type Culture Collection. Briefly, cells were maintained in Eagle's minimum essential medium supplemented with 10% FBS and 1% penicillin + streptomycin at 37°C in a humidified incubator under 5% CO₂. The cells were seeded in six-well plates at 2 × 10⁵ cells/well and randomly divided into four groups: Control, LPS (1 µg/mL) + IFN-γ (500 ng/mL), LPS (1 µg/mL) + IFN-γ (500 ng/mL) + Apigenin (20 µM), and LPS (1 µg/mL) + IFN-γ (500 ng/mL) + Apigenin (40 µM). After seeding for 24 hours, Apigenin was added, and the cells were pretreated for one hour. As followed, LPS and IFN-γ were added, and cells were collected after 24 hours. 

Cell Viability Assay

Cell viability was measured according to the instructions of Cell Counting Kit-8 (Dojindo Molecular Technologies, Inc., Kumamoto, Japan). Briefly, HMC3 cells were seeded in 96-well plates at 2000 cells/well for 24 hours. After Apigenin treatment and LPS+ IFN-γ addition for 24 hours, CCK8 solution was added, and the HMC3 cells were incubated at 37°C for 1.5 hours. The absorbances were detected at 450 nm in a microplate reader (Thermo Fisher Scientific, Waltham, MA, USA). 

Wound Healing Assay

The HMC3 cells were seeded in six-well plates at 2 × 10⁵ cells/well and incubated at 37°C under 5% CO₂ until the cells reached 80% to 90% confluence. Then the cells were wounded with a sterile micropipette tip and photographed immediately. After that, the four groups were treated by Apigenin or LPS+ IFN-γ as mentioned above. And 24 hours later, the cells were photographed again. Image J software was used to analyze the migration degree, and the result was presented as a ratio of (0–24 hours)/0 hour boundary distance. 

Transwell Assay

The Corning Transwell systems (Corning Inc., Corning, NY, USA) with inserts (8 mm pores) were applied. The HMC3 cells were seeded in a medium consisting of 2% FBS in the upper compartment at 2 × 10⁴ cells/well; the lower chamber consisted of 10% FBS. After 24 hours, Apigenin, LPS, and IFN-γ were added to the upper compartment. After another 24 hours, the HMC3 cells in the upper compartment were washed with PBS twice and fixed with 4% paraformaldehyde for 10 minutes and stained with 0.1% crystal violet (Beyotime Institute of Biotechnology) for 20 minutes. The cells on the upper surface were removed, and the remaining cells on the lower surface were photographed. ImageJ software was used to quantify the stained cells on the lower surface. 

Total RNA Extraction and Real-Time Quantitative PCR

Total RNA was extracted using Total RNA Extraction Reagent (Mei5bio, Beijing, China) according to the manufacturer's protocol. The isolated RNAs were reversed by RT Master Mix for qPCR (gDNA digester plus) (MedChemExpress, Monmouth Junction, NJ, USA). Real-time quantitative PCR (RT-qPCR) was performed using SYBR Green qPCR Master Mix (Low ROX) (MedChemExpress) and detected by an Applied Biosystems 7500 Real-Time PCR System (Thermo Fisher Scientific). The genes associated with inflammation were normalized to β-actin, and the sequences of primers are listed in the Table. 

Table.

 

Primers for RT-qPCR

 

Western Blotting

The total proteins from the whole retina tissues or HMC3 cells were extracted after homogenization with Radio Immunoprecipitation Assay (RIPA) buffer (Beyotime Institute of Biotechnology) containing 1% protease inhibitors (Beyotime Institute of Biotechnology). The protein concentrations was measured using a BCA Protein Assay-Kit (Beyotime Institute of Biotechnology). Equal amounts of protein were separated by electrophoresis and transferred to polyvinylidene difluoride membranes (Millipore, Burlington, MA, USA). The membranes were subsequently blocked with Tris-buffered saline solution with 0.1% Tween 20 (TBS-T) containing 5% (w/v) fat-free milk for two hours at room temperature. Then, membranes were incubated with primary antibodies overnight at 4°C. The next day, the membrane was washed three times with TBS-T and incubated with secondary antibodies at room temperature for one hour. The membranes were visualized by using an ECL kit (MedChemExpress), and band density was quantified using Image J software. The primary antibodies include anti-β-actin (1:10000; Affinity Biosciences, Changzhou, China), anti-TNF-α (1:1000; Abcam), anti-IL-1β (1:1000; Abcam), anti-iNOS (1:1000; Proteintech, Wuhan, China), anti-IL-6 (1:500; Affinity Biosciences), anti-Occludin (1:1000; Proteintech), anti-CD74 (1:5000; Abcam), anti-ARG1 (1:1000; Proteintech), anti-CD86 (1:1000; Proteintech), anti-TLR4 (1:1000; Abcam), anti-MyD88 (1:1000; Proteintech), anti-Nrf2 (1:1000; Proteintech), anti-HO-1 (1:1000; Proteintech). The secondary antibodies include goat anti-rabbit IgG (H+L) HRP (1:10000; Affinity Biosciences), and goat anti-mouse IgG (H+L) HRP (1:10,000; Affinity Biosciences). 

Statistical Analysis

All statistics were performed using SPSS software 26.0 (IBM, Armonk, NY, USA). For clinical and pathological scores, the Mann-Whitney U test or Kruskal-Wallis test was applied. The rest experiments were performed by one-way ANOVA followed by a least significant difference comparison test. The statistical figures were made using Prism version 8.0 software (GraphPad, San Diego, CA, USA). Significance levels are marked as follows in the figures: *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. All experiments were independently repeated at least three times. 

Results

Apigenin Treatment Ameliorated the Manifestations in EAU

To investigate whether Apigenin has therapeutic effect on autoimmune uveitis, we induced EAU and subsequently treated the mice intraperitoneally with Apigenin or vehicle daily from d7 to d13 post immunization. Dynamic observation of anterior segments by slip lamp was conducted from day 7 to day 14. As shown in Figure 1A, the EAU group showed severe conjunctival, ciliary hyperemia, anterior chamber inflammation, and synechiae. In contrast, the EAU + API group showed milder manifestation after four days of treatment (Figs. 1A, 1B). For the funduscopic assessment, the EAU group showed severe vasculitis and damage of retina and choroid. However, the EAU + API group showed mild vasculitis, which was not apparent (Figs. 1C, 1D). Moreover, the H&E staining of the EAU group revealed severe retinal fold and detachment with diffuse infiltration of the inflammatory cells, whereas the EAU + API group showed slight structural disruption of the retina and less inflammatory cell infiltration (Figs. 1E, 1F). Western blotting showed that the protein expression levels of retinal inflammatory cytokines were elevated after immunization, and Apigenin treatment reversed these changes (Fig. 1G). In addition, to evaluate the integrity of the BRB, we measured occludin expression, a tight junction protein that represents the integrity of inner retinal barrier.³² As shown in Figure 1H, Apigenin attenuated the disruption of the BRB in pathological microenvironment. These results indicate that Apigenin has a significant anti-inflammatory effect on EAU. 

Figure 1.

 

Apigenin ameliorates inflammation in EAU. (A) Representative pictures of slit lamp photography at d14 of each group (white arrow, conjunctival or ciliary congestion). (B) The clinical scores of anterior segments from d7 to d14 (*P < 0.05, the EAU+API group vs. the EAU group by Mann-Whitney U analysis, median ± interquartile, n = 4). (C) Representative fundus images of posterior pole at d14 of each group. (D)The fundus clinical scores at d14 (*P < 0.05, **P < 0.01 vs. the EAU group by Kruskal–Wallis, median ± interquartile, n = 5). (E) Representative H&E staining sections of the retina at day 14 of each group (black arrow, retinal folding; red arrow, vasculitis; green arrow, granulomatous formation) (Scale bars: 100 µm). (F) Pathological scores at day 14 according to Caspi's criteria (*P < 0.05, **P < 0.01 vs. the EAU group by Kruskal–Wallis, median ± interquartile, n = 5). (G) Representative images of Western blotting results for inflammatory cytokines in whole retinal extracts and corresponding statistics (mean ± SD; *P < 0.05, **P < 0.01, n = 3; one-way ANOVA). (H) Representative image of Western blotting results for BRB function and corresponding statistic (mean ± SD; *P < 0.05, n = 3; one-way ANOVA).

Apigenin Attenuated Retinal Microglial M1 Polarization in Pathological Conditions

Because microglia play an important role in EAU progression, and microglial M1 polarization leads to an increase of multiple inflammatory mediators, we therefore explored whether Apigenin have an impact on microglial phenotypes. As shown in Figure 2A, the EAU group exhibited increased protein expression of the M1 marker (CD74)^33–35 and reduced the M2 marker (ARG1),³⁶ whereas Apigenin treatment markedly reduced expression of CD74 but did not reverse the decrease of ARG1. To verify immunoblots results at cellular level, immunofluorescence staining of retinal flat mounts was carried out. The percentage of CD74⁺IBA1⁺/IBA1⁺ cells was increased in the EAU group, and this percentage was reduced after the administration of Apigenin (Figs. 2B, 2C). Furthermore, the immunofluorescence staining of retinal flat mounts showed a large population of amoeboid microglia in the EAU group, which was the feature of M1 phenotype, although this change was partly reversed as more multibranched microglia emerged in the Apigenin treatment group (Fig. 3A). In the pathological state, M1-activated microglia migrate to outer retina, which would exacerbate photoreceptor death.^37,38 The immunofluorescence staining of paraffin sections showed that the population of microglia were limited in the inner retina with Apigenin treatment (Fig. 3B). These results indicate that Apigenin can regulate the microglial phenotype in EAU. 

Figure 2.

 

Apigenin inhibits microglial M1 pro-inflammatory polarization in EAU. (A) Representative images of western blotting results for microglial M1/M2 marker in whole retinal extracts and corresponding statistics (mean ± SD; *P < 0.05, **P < 0.01, n = 3; one-way ANOVA). (B) Representative immunostaining images of microglia (IBA1, red) and M1 marker (CD74, green) in retinal flat mounts, nuclei were stained with DAPI (Scale bars: 100 µm). (C) Histogram summarizing the percentage of CD74⁺IBA1⁺/IBA1⁺ cells in different groups; three fields per each sample were chosen for analysis (mean ± SD; ***P < 0.001, n = 3; one-way ANOVA).

Figure 3.

 

Apigenin reduced ameboid microglia and microglia migration in EAU mice retina. (A) Representative immunostaining images of microglia (IBA1, green) in retinal flat mounts (n = 3, scale bars: 100 µm). (B) Representative immunostaining images of microglia (IBA1, green) in retinal paraffin sections, nuclei were stained with DAPI (n = 3, scale bars: 100 µm).

Apigenin Inhibited HMC3 Cell M1 Polarization

In vitro, we validated the effect of Apigenin on microglia using human HMC3 cell line. Drug toxicity was examined at five concentrations with CCK8, and the results showed no cytotoxicity for HMC3 under both the physiological condition and the pathological condition. A previous study showed an obvious anti-inflammatory effect on BV2, a mouse-derived microglia cell line in 20 µM Apigenin.²² Therefore the two concentrations (20 µM and 40 µM) were chosen for functional experiments (Fig. 4A). As indicated in Figure 4B, LPS + IFN-γ treatment enhanced the transcription of inflammatory cytokines, and the addition of Apigenin greatly blocked it. Apigenin decreased protein expression of iNOS, TNF-α, IL-6, and IL-1β in a dose-dependent manner (Fig. 4C). Consistent with that, the level of M1 polarization was inhibited (Fig. 4D). Moreover, the wound healing and Transwell assays were conducted to assess the microglial migration ability. As shown in Figures 4E and 4F, Apigenin inhibited microglial migration. These results emphasize again that Apigenin can ameliorate microglial M1 polarization and the inflammatory response in the state of pathology. 

Apigenin Attenuated Microglia Inflammation Via TLR4/MyD88 Signaling Pathway

Last, we explored the molecular mechanisms and signaling pathways of Apigenin regulation. Because the TLR4 signaling pathway is one of the major components of proinflammatory signaling pathways and is involved in microglial activation³⁹, the protein expression of TLR4 and MyD88 were therefore detected. In comparison of normal control, the TLR4/MyD88 signaling pathway were activated in stimulation group, whereas Apigenin blocked it (Fig. 5A). Furthermore, we detected the expression of nuclear factor (erythroid-derived 2)–like 2 (Nrf2), which is a transcription factor related to M2 anti-inflammatory polarization of microglia.⁴⁰ In our study, Apigenin treatment did not up regulate the protein expression of Nrf2 and HO-1 significantly in stimulated cells (Fig. 5B). The present data suggested that Apigenin alleviated EAU, partly through its inhibition of the microglial M1 activation via TLR4/MyD88 signaling pathways. 

Figure 4.

 

Apigenin inhibits inflammatory cytokines production and M1 polarization in HMC3 cell line. (A) CCK8 assays were carried out to test drug toxicity on HMC3 cell lines both under physiological condition and pathological condition. (B) RT-qPCR to analyze mRNA level for each inflammatory cytokine. (C) Representative images of Western blotting results for inflammatory cytokines and corresponding statistics. (D) Representative images of Western blotting results for microglial M1/M2 markers and corresponding statistics. (E) Wound healing assays to test microglia mobility and corresponding statistics (scale bars: 200 µm). (F) Transwell assays to test microglia migration ability and corresponding statistics (scale bars: 200 µm). All above data show as mean ± SD; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; ns, not significant; n = 3; one-way ANOVA.

Figure 5.

 

Apigenin inhibits microglial M1 polarization via TLR4/MyD88 signaling. (A) Representative images of Western blotting results for TLR4/MyD88 signal and corresponding statistics. (B) Representative images of western blotting results for Nrf2/HO-1 signal and corresponding statistics. Above data show as mean ± SD; *P < 0.05, **P < 0.01; ns, not significant; n = 3; one-way ANOVA.

Discussion

Considering the indispensable role of microglia in various central nervous system diseases, the plasticity of microglia has become a therapeutic target for attenuating the damage of local tissue. There are multiple studies reporting that regulation of microglia polarization can reduce the severity of EAU.^17,39 The modulation of microglia activation by Apigenin has been reported in BV2 and MG5 cell lines; in this report Apigenin effectively reduced the production of inflammatory factors and phagocytosis of microglia.²⁴ This time, through experiments, we identified a powerful therapeutic effect of Apigenin on a model of human autoimmune uveitis, a rodent model induced by hIRBP. It is known that day 14 to day 21 after immunization is the “early phase” of EAU and the peak of inflammation, and more than 21 days is the “late phase” of EAU during which the inflammation is inclined to decline.¹⁶ Here, we chose day 14 as an observation time point to assess the efficacy of Apigenin. We administered Apigenin intraperitoneally at a dose of 20 mg/kg from day 7 to day 13, and then we found that Apigenin reduced inflammatory cytokines in the retina, accompanied by a reduction of M1 proinflammation polarization of the microglia. We found that Apigenin reduced the expression of CD74; however, it did not significantly elevate the expression of ARG1, a M2 marker related to the homeostasis and repair, indicating that Apigenin can inhibit microglial M1 polarization, but it may not achieve M1/M2 phenotype transformation at day 14. To detect whether we could yield insightful data regarding the appearance of M2 microglia, we collected mouse retinal protein on day 21 and found that the ARG1 level was increased significantly after Apigenin's therapeutic window was prolonged to day 21 or changed to day 14 to day 21 (Supplementary Figs. S1B, S1C). Furthermore, similar results were obtained from the withdrawal group in which Apigenin treatment stopped at day 14 (Supplementary Fig. S1A). These results indicated that the recovery stage might come earlier in the presence of Apigenin and the microglia would switch to the M2 phenotype to promote tissue repair and shorten the progression of disease. And Apigenin may play a different beneficial role in the different course of disease to relieve inflammation. Besides, Apigenin can still promote the M2 microglia conversion after withdrawal, and this may be because Apigenin has a slow metabolism and a slow elimination phase^41,42; thus it would accumulate in the body and the effect would persist. 

In vitro, we verified the effect of Apigenin using the HMC3 human microglia cell line. We also found that Apigenin blocked microglial M1 polarization but failed to reprogram microglia to the M2 phenotype, which is consistent with the previous findings in BV2. Finally, we focused on two signaling pathways associated with microglia phenotype modification. Nrf2 is an antioxidant transcription factor that is a key regulator in the disorders related to oxidative stress, and it is a shared target of various flavonoids.^43,44 It also plays an important role in maintaining intracellular homeostasis in the microglia, protecting the cell from overactivation and organelle damage caused by an accumulation of reactive oxygen species. Surprisingly, Apigenin did not increase Nrf2/HO-1 protein expression in our study. We also found that Apigenin significantly blocked the TLR4/MyD88 pathway, a pathway that was seen in EAU and microglial polarization. Apigenin decreased TLR4/MyD88 in a dose-dependent manner consistent with the changes of the expression of inflammatory factors and polarization markers, suggesting the nonselective action of Apigenin, which is similar to the characteristics of many natural products, and that apigenin might tend to inhibit some proinflammatory factors rather than selectively activating anti-inflammatory factors. 

Some limitations need to be elucidated: like other flavonoids, Apigenin has poor solubility in water and fat; therefore its bioavailability is limited. Some drug chemical modifications have been reported.⁴⁵ A solid dispersion of Apigenin has been used in dry AMD, and its bioavailability is better than original Apigenin.²³ We believe that if this improved drug is adopted in EAU, it may reduce the inflammatory manifestations more efficiently. In future experiments, we may use this solid dispersion to explore more of the therapeutic potential of Apigenin in other ocular diseases. Additionally, although multiple studies have used microglia stimulated by LPS and IFN-γ to simulate EAU in vivo,^17,29 further studies are needed to find better stimulating methods of microglia to simulate EAU model. 

In conclusion, we demonstrated that Apigenin is able to alleviate autoimmune uveitis by modulating the polarization of retinal microglia. Apigenin influences polarization of HMC3 via the TLR4/MyD88 pathway. 

Acknowledgments

Supported by the National Natural Science Foundation Project of China (82070951, 82271078), the National Key Clinical Specialties Construction Program of China, Chongqing Branch of the National Clinical Research Center for Ocular Diseases, Chongqing Key Laboratory of Ophthalmology (CSTC, 2008CA5003), and the Program for Youth Innovation in Future Medicine, Chongqing Medical University (w0047). 

Disclosure: N. Shu, None; Z. Zhang, None; X. Wang, None; R. Li, None; W. Li, None; X. Liu, None; Q. Zhang, None; Z. Jiang, None; L. Tao, None; L. Zhang, None; S. Hou, None