The DNA Methylation Inhibitor Zebularine Controls CD4+ T Cell Mediated Intraocular Inflammation

CD4+ T cell mediated uveitis is conventionally treated with systemic immunosuppressive agents, including corticosteroids and biologics targeting key inflammatory cytokines. However, their long-term utility is limited due to various side effects. Here, we investigated whether DNA methylation inhibitor zebularine can target CD4+ T cells and control intraocular inflammation. Our results showed that zebularine restrained the expression of inflammatory cytokines IFN-γ and IL-17 in both human and murine CD4+ T cells in vitro. Importantly, it also significantly alleviated intraocular inflammation and retinal tissue damage in the murine experimental autoimmune uveitis (EAU) model in vivo, suggesting that the DNA methylation inhibitor zebularine is a candidate new therapeutic agent for uveitis.


INTRODUCTION
Autoimmune uveitis is a heterogeneous collection of diseases characterized by intraocular inflammation (1). Affected patients are at risk of visual impairment or blindness (2), in particular those who are resistant or intolerant to conventional immunosuppressive therapies (3)(4)(5). Consequently there is a need to develop improved approaches to achieve control of intraocular inflammation.
Abnormal activation of the T helper (Th) cells and imbalance between inflammatory Th1/Th17 and regulatory T (Treg) cells play a key role in the pathogenesis of autoimmune uveitis (6)(7)(8). Modulating Th cell differentiation and function has been proposed as a therapeutic strategy for uveitis (9). The expression of signature cytokines and master transcription factors, as well as Th functions, are under tight control of coordinated epigenetic alterations (10,11). And many genome-wide and locus-specific epigenetic changes have been found in patients with immune-mediated diseases such as systemic lupus erythematosus and rheumatoid arthritis (12). Therefore, modulating the epigenetic program that controls Th1, Th17, and Treg functions may serve as a new way to control the intraocular inflammation in uveitis.
Two key epigenetic mechanisms-DNA methylation and histone modifications-regulate chromatin accessibility, and have been the targets of a series of compounds developed in recent years for the treatment of cancers and immune-mediated diseases (12). Among these small-molecule inhibitors, two DNA methyltransferase (DNMT) inhibitor pro-drugs, 5-azacytidine and 2 ′ -deoxy-5-azacytidine have been approved for the treatment of myelodysplastic syndrome and acute myeloid leukemia (13). In addition, an anti-inflammatory effect of these two drugs has also been observed in murine models of asthma (14) and experimental autoimmune encephalomyelitis (15). We therefore hypothesize that another DNA methylation inhibitor, zebularine, which has relatively low cellular toxicity and a longer half-life (16), has the potential to suppress uveitis.
Here, to elucidate whether zebularine is able to control the intraocular inflammation through regulating the expression of inflammatory cytokines in CD4 + T cells, we explored the changes of IFN-γ, IL-17, and Foxp3 (17) in response to zebularine treatment in Th1, Th17, and Treg cells. In addition, the immunosuppressive effects of zebularine in vivo were also evaluated using the murine experimental autoimmune uveitis model.

MATERIALS AND METHODS
Human Peripheral Blood CD4 + T Cell Isolation CD4 + T cells were obtained by negative selection from peripheral blood of healthy controls (HCs) (N = 4; 3 female and 1 male; average age of 42) following informed consent in accordance with National Health Service Research Ethic Committee approved protocols at the University Hospitals Bristol Foundation Trust, United Kingdom (04/Q2002/84). Written, informed consent was obtained from all study participants. CD4 + T cells were obtained by incubating up to 80 ml uncoagulated peripheral blood with RosetteSep TM Human CD4 + T cell Enrichment Cocktail (Stemcell Technologies, Canada) according to the manufacturer's instructions. Blood was then layered on Ficoll-Paque PLUS (GE Healthcare, USA) and centrifuged for 1,200 × g for 20 min. Enriched CD4 + T cells were removed from the density gradient and washed in RPMI-1640 (Thermofisher) supplemented with 10% fetal calf serum (FCS). The purity of human CD4 + T cells was >95%.

Mice
Female C57BL/6 2-Hb mice (8-10 weeks) were purchased from Guangdong Medical Laboratory Animal Center (China) and maintained in a SPF facility in Zhongshan Ophthalmic Center of Sun Yat-sen University. All animal experiments were approved by the Institutional Animal Care and Use Committee of Zhongshan Ophthalmic Center. All animal work was performed in compliance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research.

RNA-seq Analysis
Total RNA from Th1, Th17, and Treg cells was extracted with the MasterPure Complete DNA and RNA Purification Kit (Epicentre, UK) according to the manufacture's instruction. A total of 100 ng RNA was sonicated into fragments of 300-400 base pairs using Bioruptor (Diagenode, Belgium). mRNA library was prepared using VAHTS mRNA-seq V3 Library Prep Kit for Illumina (Vazyme, Nanjing, China) following the manufacture's protocol and sequenced on the Illumina HiSeq2500 sequencer with HiSeq SR Cluster Kit V4 and HiSeq SBS Kit V4 50 cycle kit (Illumina). The initial processing was performed by CASAVA (v1.8.2). Sequencing reads were then subjected to quality control processed by FastQC (v0.11.5) and trimmed by Cutadapt (v1.9.1). Quality controlled reads were then analyzed using DEseq2.

Quantitative Real-Time PCR (qPCR)
Total RNA from T cells was extracted with the MasterPure Complete DNA and RNA Purification Kit (Epicentre, UK), and cDNA was synthesized with the PrimeScript TM cDNA Synthesis Kit (Takara, Japan). qRT-PCR was performed using SYBR Green PCR Master Mix (KAPA Biosystems, USA) on the Light Cycler 480 instrument (Roche, Switzerland). The samples were run in duplicate and the relative expression was determined by normalizing the expression of each gene to glyceraldehyde 3-phosphate dehydrogenase (GAPDH) using the 2 − Ct method. The primers used are shown in the Supplementary Table 1.

Drug Treatment in Mice
Zebularine was reconstituted with 0.9% saline to make a 1 µg/µL stock and stored at −80 • C. Zebularine was administered to the mice at 10 µg/g body weight by intraperitoneal injection. The injection was started on day 7 post-immunization and performed daily for 7 consecutive days. The control group was injected with 0.9% saline only. Body weights of both groups were monitored after drug treatment.

EAU Induction and Scoring
C57BL/6 mice were immunized subcutaneously, at the base of tail (100 µL) and in both thighs (50 µL), with total 50 µg human interphotoreceptor retinoid binding (IRBP) 651−670 emulsified with complete Freund's adjuvant supplemented with 3.5 mg Mycobacterium tuberculosis. Each mouse also received 1 µg pertussis toxin (Tocris Bioscience, UK) intraperitoneally. To assess the clinical score of EAU, we performed dilated-pupil fundus examination with a Micron III murine fundus camera (Phoenix Research Labs, USA) and histological assessment of FFPE sections of the retinal tissues from EAU mice (18). The clinical scores were given by two independent experienced observers in a blindfold manner based on the criteria for EAU scoring as described previously (19).

Bisulfite Sequencing
Total DNA isolated from mouse polarized Treg cells was extracted with the MasterPure Complete DNA and RNA Purification Kit (Epicentre, UK). Sodium bisulfite treatment of total DNA was performed using the EZ DNA Methylation-Gold Kit (Zymo Research, USA) according to the manufacturer's instruction. PCR amplification was performed on the T100 Thermal Cycler (Bio-Rad, USA) in a final volume of 50 µL. PCR products were purified with the Zymoclean Gel DNA Recovery Kit (Zymo Research, USA). The sequencing results were analyzed by BiQ Analyzer (Max Planck Institut Informatik, Germany). The following primers pairs were used: Foxp3 enhancer CpG

Statistical Analysis
Statistical analysis was performed with Prism Graphpad 7.0 (GraphPad Software, USA). Mann-Whitney U test or one-way ANOVA test was used accordingly.

Zebularine Restrains the Expression of IFN-γ and IL-17A in CD4 + T Cells in vitro
To assess the effects of zebularine on the expression of proinflammatory cytokines, we first examined the human peripheral CD4 + T cells. FACS-sorted CD4 + CCR6 − and CD4 + CCR6 + cells were stimulated with anti-CD3/anti-CD28 antibodies or under the Th17 polarizing condition, respectively. As shown in Figure 1, the frequency of IFN-γ + IL-17 − cells was significantly reduced in both CCR6 + and CCR6 − cells in response to zebularine treatment (Figures 1A,B), while IL-17 expression was restrained by zebularine in a dose dependent manner (Figures 1A,C,D). The viability of CD4 + T cells was not significantly affected by zebularine treatment (Figure 1E). These data suggest that zebularine is able to control the inflammatory cytokine expression in human CD4 + T cells.
To further explore the immunomodulatory effect of zebularine in murine cells, we differentiated murine peripheral CD4 + T cells into inflammatory Th1 and Th17 conditions. Similar to the findings in human CD4 + T cells, zebularine was found to significantly suppress the protein expression of IFN-γ and IL-17A (Figures 2A,B) in murine Th1 and Th17 cells, respectively. However, the cell viability was significantly compromised in response to those high dose zebularine treatments (Figure 2C). To confirm the global changes zebularine stimulation may lead to on CD4 + T cells, we performed genome-wide expression analysis of murine Th1 and Th17 cells in response to zebularine stimulation using the mRNA-seq technology. The principal component analysis ( Figure 3A) and hierarchical clustering analysis (Figure 3B) of all of the differentially expressed genes (defined as a two-fold change between any two conditions, with a one-way ANOVA  Figure 3C). We next performed quantitative PCR analysis to confirm the changes of mRNA expression of the inflammatory cytokines and key transcription factor Tbx21, which encodes the Th1 master regulator Tbet found by RNA-seq analysis. The expression of Ifng (Figure 4A), Tbx21 (Figure 4C), and Il17a ( Figure 4B) was significantly decreased by zebularine in Th1 and Th17 cells, respectively, while the relative RNA expression of Rorc was increased in Th17 cells in response to zebularine treatment ( Figure 4D). Taken together, our data demonstrated the immunomodulatory function of zebularine on both human and murine CD4 + T cells in vitro.

Zebularine Promotes the Expression of Foxp3 in Murine CD4 + T Cells in vitro
In addition to the effects on cytokine expression, we also investigated whether zebularine modulated the expression of Foxp3, the key transcription factor controlling the development and function of Treg (20). The genome-wide expression analysis demonstrated a significant response of Treg cells to the zebularine stimulation (Figures 3A,B). As shown in Figures 3C, 5A,B, the RNA expression of Foxp3 was activated and significantly increased in response to the treatment of zebularine in Th1 and Th17 cells. Under inducible Treg polarizing condition, high dose of zebularine was not able to further elevate the expression of Foxp3 (Figures 5C-E). It is noteworthy that zebularine treatment at 10 µg/mL significantly decreased the viability of Th1 cells (Figure 2C), while the Treg cells were still highly viable ( Figure 5F). These data suggest that zebularine promotes the expression of Foxp3 in CD4 + T cells.
The epigenetic regulation of Foxp3 expression by DNA methylation alterations was well documented by previous studies (21). Therefore, we next assessed whether the methylation status of the CpG islands located in Foxp3 enhancer was changed by zebularine in Treg cells, using the bisulfite sequencing technique. As shown in Figure 5G, the Foxp3 upstream enhancer (−5786 to −5558) was hypermethylated in the vehicle-treated control cells (66% methylated), while all 23 CpG islands were demethylated in cells treated with zebularine (10 µg/mL). These data further demonstrate that zebularine promotes the expression of Foxp3 in CD4 + T cells through demethylating its enhancer region.

Zebularine Controls Intraocular Inflammation in vivo
To investigate the anti-inflammatory effect of zebularine in vivo, we treated EAU mice intraperitoneally with zebularine and investigated the changes of intraocular inflammation. Fundus examination was performed every 7 days to evaluate the severity of intraocular inflammation in EAU mice. As shown in Figures 6A,B, the retinal tissue damage started on Day 7 and peaked at Day 14 post immunization. However, zebularine treatment significantly reduced the severity of intraocular inflammation and retinal tissue damage, evidenced by the fundus clinical score reduction from 2.91 ± 0.24 (vehicle treated group) to 0.88 ± 0.16 (zebularine treated group) on Day 14. Importantly, during the 5 weeks when mice received zebularine treatment, we did not observe any body weight changes ( Figure 6C). Consistent with the fundus examination result, histological analyses of the retinal tissues from EAU mice on the disease peak Day 14 post immunization revealed that zebularine treatment significantly alleviated intraocular inflammation and retinal tissue damage (Figures 7A-C). Taken together, our data suggest that zebularine significantly reduces intraocular inflammation and retinal damage in vivo.
Since zebularine restrained inflammatory cytokine expression in CD4 + T cells in vitro, we next prepared single cell suspensions from the eyes, cervical lymph nodes (CLN), peripheral lymph nodes (PLN) and spleen of EAU mice treated with or without zebularine on Day 14 post-immunization and analyzed the changes of IFN-γ, IL-17, and Foxp3 expression. As shown in Figures 8A-E, the frequency of both intraocular CD4 + IFN-γ + and CD4 + IL-17 + cells was reduced by zebularine, while the frequency of CD4 + Foxp3 + cells was elevated (Figures 8B,F). In contrast, the expression of IFN-γ, IL-17, and Foxp3 was not significantly changed in CLN, PLN, and spleen (Figures 8A-F). Therefore, these data further demonstrate that zebularine controls ocular specific inflammation without significantly altering the systemic CD4 + T cell populations.

DISCUSSION
CD4 + T cells mediate systemic and local inflammation in uveitis (22). Our current study suggests that the DNA methylation inhibitor zebularine suppresses the expression of inflammatory cytokines IFN-γ and IL-17 in CD4 + T cells in vitro and in vivo, promotes Foxp3 expression, and alleviates the severity of intraocular inflammation and retinal tissue damage in EAU mice. To our knowledge, this is the first report showing that zebularine is a potential therapeutic agent for uveitis.
Previous studies have demonstrated the clinical utility of DNA methylation inhibitors in oncology, especially in the treatment of myelodysplastic syndrome and acute myeloid leukemia (12). However, it is only recently that these drugs have been evaluated in the context of experimental models of inflammatory disease (14,15,(23)(24)(25)(26)(27). This showed that systemic administration of the two DNA methylation inhibitors azacytidine and decitabine for treating immune-mediated diseases was limited due to their significant cytotoxicity and side-effects (28). In contrast, our data revealed limited cytotoxicity of zebularine on human CD4 + T cells in vitro and limited systemic toxicity of zebularine in mice in vivo, suggesting that zebularine may be a better immunosuppressive agent in comparison to the other DNA methylation inhibitors. Importantly, several reports have identified indoleamine 2,3-dioxygenase (IDO) as the targets of zebularine through which zebularine carried out immunomodulatory functions on cancer and autoimmunity treatments (29)(30)(31). Our data on CD4 + T cells provided another possible mechanism by which zebularine carried out its immunosuppressive function.
Tbet, RORγt, and Foxp3 are the canonical transcription factors promoting Th1, Th17, and Treg polarization (11). We found that zebularine suppressed the expression of Tbx21 (encoding Tbet), but promoted the expression of Foxp3 and Rorc (encoding RORγt). Suppression of Tbet in Th1 cells correlated well with the significant reduction of IFN-γ in response to zebularine treatment. Promotion of Foxp3 expression, which is consistent with previous reports that other DNA methylation inhibitors also activate Foxp3 expression in murine models of asthma, EAE and diabetes (14,24,32,33), explained the significant alleviation of intraocular inflammation found in EAU mice. However, upregulation of Rorc expression seemed contradictory to the reduction of IL-17 expression by zebularine found both in vitro and in vivo. Although RORγt is the critical transcription factor promoting IL-17 expression and Th17 differentiation, other transcription factors such as STAT3, IRF4, and BATF are also responsible for the coordinated regulation of IL-17 expression (34). Therefore, the suppression of IL-17 expression by zebularine may depend on a RORγt independent pathway.
Our study demonstrates that zebularine restrains IFN-γ and IL-17 expression and promotes Foxp3 expression in CD4 + T cells and may serve as a candidate therapeutic agent for autoimmune uveitis. Further study is warranted to elucidate the molecular mechanisms by which zebularine epigenetically regulates the expression of these key cytokines and transcription factors in T cells.

DATA AVAILABILITY
The datasets generated for this study are available on request to the corresponding author.

ETHICS STATEMENT
CD4 + T cells were obtained from peripheral blood from healthy controls (HCs) following informed consent in accordance with National Health Service Research Ethic Committee approved protocols at the University Hospitals Bristol Foundation Trust, United Kingdom (04/Q2002/84). Written informed consent was obtained from all study participants. All animal experiments were approved by the Institutional Animal Care and Use Committee of Zhongshan Ophthalmic Center. All animal work was performed in compliance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research.