Interleukin-17A Induces IL-1B Secretion from RPE Cells Via the NLRP3 Inflammasome

Retinal Cell Biology Interleukin-17A Induces IL-1b Secretion From RPE Cells Via the NLRP3 Inﬂammasome

Shujie Zhang,1–3 Ning Yu,4 Rong Zhang,1–3 Shenghai Zhang,1–3 and Jihong Wu1–3

1Research Center, Eye & ENT Hospital of Fudan University, Shanghai, China 2Key Laboratory of Myopia, Ministry of Health, Fudan University, Shanghai, China 3Shanghai Key Laboratory of Visual Impairment and Restoration, Fudan University, Shanghai, China 4Department of Dermatology, Shanghai Skin Disease Hospital, Shanghai, China

Correspondence: Jihong Wu, Re- PURPOSE. Inflammasome activation and IL-1b production have been proposed to have an search Center, Eye & ENT Hospital, important role in age-related macular degeneration (AMD). Growing evidence is emerging for Shanghai Medical College, Fudan involvement of interleukin-17A (IL-17A) in AMD pathogenesis. We investigated the effects of IL- University,83FenyangRoad,Shang- 17A on the activation of inflammasome and production of IL-1b in primary human RPE cells. hai 200031, China; [email protected]. METHODS. Primary human RPE cells were isolated and cultured for the following experiments. Shenghai Zhang, Research Center, Expression patterns of IL-17 receptor A (IL-17RA), IL-17 receptor C (IL-17RC), and ACT1 were Eye & ENT Hospital, Shanghai Med- analyzed by RT-PCR, flow cytometry, and immunofluorescence. IL-17A was added to the cell ical College, Fudan University, 83 cultures, and cytokine expression, signaling pathways, and inflammasome machinery were Fenyang Road, Shanghai 200031, investigated using real-time RT-PCR, ELISA, Western blot, flow cytometry, and small China; interfering RNA. [email protected]. RESULTS. Retinal pigment epithelial cells constitutively expressed IL-17RA, IL-17RC, and ACT1. SZ and NY contributed equally to the IL-17A upregulated the mRNA levels of pro-IL-1b, IL-8, CCL2, and CCL20, as well as the work presented here and should protein level of IL-1b. IL-17A induced the phosphorylation of Akt, Erk1/2, p38 MAPK, and NF- therefore be regarded as equivalent jB p65 in RPE cells. Blocking NF-jB attenuated IL-17A–induced expression of pro-IL-1b authors. mRNA. IL-17A enhanced pro-caspase-1 and NLRP3 mRNA expression. Inhibiting caspase-1 Submitted: June 27, 2015 activity and silencing NLRP3 decreased IL-1b secretion, confirming NLRP3 as the IL-17A– Accepted: December 8, 2015 responsive inflammasome on the posttranscriptional level. The mechanism of IL-17A– triggered NLRP3 activation and subsequent IL-1b secretion was found to involve the Citation: Zhang S, Yu N, Zhang R, Zhang S, Wu J. Interleukin-17A induc- generation of reactive oxygen species. es IL-1b secretion from RPE cells via CONCLUSIONS. Our results suggest that IL-17A triggers a key inflammatory mediator, IL-1b, from the NLRP3 inflammasome. Invest RPE cells, via NLRP3 inflammasome activation, holding therapeutic potential for AMD. Ophthalmol Vis Sci. 2016;57:312– 319. DOI:10.1167/iovs.15-17578 Keywords: RPE cells, IL-17A, inflammasome, AMD

ge-related macular degeneration (AMD) is a neurodegener- it still is not clear whether this component of innate immunity A ative disease with a multifactorial etiology, primarily could be activated or regulated by the local inflammatory attributable to morphologic and functional abnormalities in cytokine milieu generated during adaptive immune responses. the RPE cells.1 Mounting evidence has highlighted the essential Interleukin-17A (IL-17A) is the signature cytokine of Th17 role of immune processes in the development, progression, and cells and has a pivotal role in inducing expression of treatment of AMD. Innate and adaptive immune systems have proinflammatory cytokines and chemokines in the pathogenesis been shown to contribute to AMD pathogenesis.2,3 of autoimmune and inflammatory diseases.16 Recent studies Inflammasome activation is a key component of innate have demonstrated involvement of IL-17A in the pathogenesis of immunity.4 Recent efforts have drawn attention to the AMD. Aberrant levels of IL-17A were observed in AMD macular inflammasome machinery, particularly its product IL-1b,asan lesions and sera.17 Complement component C5a, which inciting factor in AMD.5–8 It has been well documented that accumulated in AMD patient serum and in drusen, induced IL- RPE cells can be triggered to secrete a number of inflammatory 17A production from CD4þ T cells.18 Arecentepigeneticstudy cytokines and chemokines.9 Among these, IL-1b is one of the demonstrated that hypomethylation of the IL-17 receptor C (IL- initial cytokines produced following RPE activation.10,11 Tran- 17RC) promoter was associated with AMD.19 Since previous scription of the pro-IL-1b gene and production of cytosolic pro- reports showed the presence of IL-17A receptors on RPE IL-1b are dependent on the activation of NF-jB. At the cells,20,21 possibility exists that IL-17A may have a role in AMD posttranscriptional level, generation of mature IL-1b is regulat- by direct modulating the immunologic functions of RPE cells. ed by inflammasomes, which is formed by NACHT, LRR, and Both IL-17A signaling and inflammasome machinery have PYD domains-containing (NLRP) proteins, the most studied been linked to the pathogenesis of AMD; however, to our member of which is NLRP3.12 More recently, the role of the knowledge no studies to date have adequately addressed the NLRP3 inflammasome in AMD pathogenesis has been investi- interactions between them. Interleukin-1b functions in synergy gated extensively using AMD-related stimuli, such as A2E,8 Alu with IL-23 to promote the production of IL-17A and related RNA,13 lysosomal activation,14 and oxidative stress.15 However, cytokines from Th17 cells.22,23 Therefore, we hypothesized that

iovs.arvojournals.org j ISSN: 1552-5783 312

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. Downloaded from tvst.arvojournals.org on 10/01/2021 IL-17A Induces IL-1b Secretion From RPE Cells IOVS j February 2016 j Vol. 57 j No. 2 j 313

TABLE. The Sequence of Primers Used for RT-PCR and Real-Time RT- removed, the posterior poles were incubated in 2% dispase PCR solution (Gibco, Grand Island, NY, USA) for 30 minutes in 378C, 5% CO2. After dispase treatment, the posterior poles were Gene Primers dissected into quadrants, and the retina was removed gently with IL-17RA 50-CGT TTG TGC GTC AGG TTT G-30 forceps. Single-cell RPE layers were peeled off in small sheets, 50-GCA GGA TCT GGT AAT GGG TAG A-30 and then RPE cells were dissociated from the sheets in a solution IL-17RC 50-AGA TGC CTG TGC CCT GGT TCT TGC T-30 containing 0.25% trypsin and 0.02% ethylenediaminetetraacetic 50-AGC AAG AAC CAG GGC ACA GGC ATC T-30 acid (Life Technologies, Grand Island, NY, USA) at 378Cfor30 ACT1 50-CCG TGA TGA TAA TCG TAG C-30 minutes. Afterward, the enzyme-treated tissues were dissociated 50-TAG CAT TTG GGA AGA GCA-30 into single RPE cells by gentle pipetting and centrifugation. Pro-caspase-1 50-TGA AGG ACA AAC CGA AGG-30 The RPE cells were cultured in DMEM:F12 growth medium 50-GAA GAG CAG AAA GCG ATA-30 containing 5.5 mM glucose (Gibco), supplemented with 10% NLRP3 50-GCC TCA ACA AAC GCT ACA-30 fetal bovine serum (Gibco). The cell cultures were maintained 50-GAC GCC CAG TCC AAC ATC-30 in a humidified incubator at 378C in an atmosphere containing Pro-IL-1b 50-AGC TAC GAA TCT CCG ACC AC-30 95% air and 5% CO2. The cells were passed every 3 days by 50-CGT TAT CCC ATG TGT CGA AGA A-30 trypsinization. For the following experiments, RPE cells were IL-8 50-ACT CCA AAC CTT TCC ACC-30 seeded into 6-well plates (Corning, Lowell, MA, USA). The 50% 50-CTT CTC CAC AAC CCT CTG-30 confluent RPE cell cultures were used in RPE65 immunoflu- CCL2 50-CTT CTG TGC CTG CTG CTC-30 orescence and the 80% to 100% confluent RPE cell cultures at 50-TGC TGC TGG TGA TTC TTC T-30 passages 3 to 5 were used in other experiments. CCL20 50-TGC TGC TAC TCC ACC TCT-30 After treatment with IL-17A (up to 200 ng/mL), Bay 11-7082 50-GCA AGT GAA ACC TCC AAC-30 (up to 10 lM), Z-YVAD-fmk (up to 10 lg/mL), small interfering GAPDH 50-AAT CCC ATC ACC ATC TTC C-30 RNA (siRNA), or NAC (5 mM), RPE cells were tested for viability 50-TTG AGG CTG TTG TCA TAC TTC T-30 using LIVE/DEAD Fixable Near-IR Dead Cell Stain Kit (Invitro- gen, Carlsbad, CA, USA) and flow cytometry. The cell viability was above 90% and did not decrease following these treatments. local IL-17A may serve as a trigger for inflammasome signaling and IL-1b production in RPE cells, forming a positive feedback Immunofluorescence of RPE Cells loop. In the present study, we first aimed to examine the expression pattern of IL-17 receptor A (IL-17RA), IL-17RC, and The 50% confluent cultured RPE cells were fixed in PBS ACT1 in primary human RPE cells. We next characterized the containing 3.7% paraformaldehyde. They then were blocked putative role of IL-17A in the production of IL-1b and the with the appropriate normal serum in PBS for 1 hour at room activation of NLRP3 inflammasome machinery in RPE cells. temperature. They then were incubated at 48C overnight with anti-RPE65 (1:200; Abcam). After being washed in PBS, the sections were incubated further with FITC-conjugated anti- MATERIALS AND METHODS mouse IgG (eBioscience) for 45 minutes at 378C. Control sections were treated in the same way, but with no primary Reagents mouse anti-RPE65 monoclonal antibody. Cell nuclei were counterstained with Hoechst 33342 (Life Technologies). Bay 11-7082 was purchased from Cell Signaling Technology (Danvers, MA, USA). Caspase-1 inhibitor (Z-YVAD-fmk) was obtained from Biovision (Milpitas, CA, USA). N-acetyl-L-cysteine Immunofluorescence of Human Retina (NAC) was obtained from Sigma-Aldrich Corp. (St. Louis, MO, Surgically isolated pieces of human retina-RPE-choroid were USA). For flow cytometry, anti-human IL-17RA, anti-human IL- fixed in 4% paraformaldehyde at 48C overnight and then washed 17RC, and appropriate IgG1 isotype antibody were purchased in cold PBS for 20 minutes. Tissues were cryoprotected before from R&D Systems, Inc. (Minneapolis, MN, USA); anti-human freezing by overnight incubation in 20% sucrose solutions in PBS ACT1 was obtained from eBioscience (San Diego, CA, USA). and then in 30% sucrose in PBS overnight at 48C. Tissues were For Western blot, antibodies against phospho-Akt, total-Akt, embedded in optimal cutting temperature embedding medium phospho-Erk1/2, total-Erk1/2, phospho-p38 MAPK, total-p38 (Sakura Finetek, Torrance, CA, USA) and frozen in liquid MAPK, phospho-NF-jBp65,b-actin, and caspase-1 were nitrogen. Cryosections (1–15 lm) were cut and collected on purchased from Cell Signaling Technology. Anti–total-NF-jB glass slides, dried at room temperature, and stored at808C until p65 and anti-human NLRP3 were obtained from Abcam use. The retina cryosections were washed with PBS and (Cambridge, MA, USA). Secondary antibodies were horseradish incubated with blocking solution (10% goat serum in PBS) at peroxidase-conjugated anti-rabbit, and anti-mouse IgG (Cell room temperature for 2 hours and then with goat anti–IL-17RA Signaling Technology). (1:200 dilution, sc-23124; Santa Cruz Biotechnology, Inc., Dallas, TX, USA) or rabbit anti–IL-17RC (1:200 dilution, sc-367206; RPE Cells Isolation and Culture Santa Cruz Biotechnology, Inc.) at 48C overnight. After they were washed with PBS three times, sections were incubated at The study was approved by the ethics committee of the EYE & room temperature for 1 hour with secondary antibodies coupled ENT Hospital of Fudan University, and was carried out in to Alexa Fluor 488 (1:1000 dilution; Invitrogen) and Alexa Fluor accordance with the ethical principles in the Declaration of 555 (1:1000 dilution; Invitrogen), respectively. Nuclei were Helsinki. Informed consents concerning the use of the posterior counterstained with Hoechst 33342. Sliders were analyzed on a tissues of the donated eyes for research purposes were obtained scanning laser confocal microscope. by the Eye Bank of the EYE & ENT Hospital of Fudan University. Primary RPE cells were isolated within 24 hours of death from RT-PCR and Real-Time RT-PCR healthy adult human globes provided by the Eye Bank of the EYE & ENT Hospital of Fudan University.24 On receipt, intact globes Total RNA was extracted from RPE cells using RNeasy Mini Kit were rinsed in PBS. After the anterior portion of the eye was (Qiagen, Valencia, CA, USA) in accordance with the manufac-

Downloaded from tvst.arvojournals.org on 10/01/2021 IL-17A Induces IL-1b Secretion From RPE Cells IOVS j February 2016 j Vol. 57 j No. 2 j 314

FIGURE 1. Expression of IL-17RA, IL-17RC, and ACT1 in RPE cells. (A) Cultured RPE cells morphologic analysis. Left: On the seventh day of the primary culture, RPE cells contained some pigment. Right: Positive RPE65 staining, indicated by the green color, was detected in RPE cells. (B) The mRNA expression of IL-17RA, IL-17RC, and ACT1 in cultured RPE cells were analyzed by RT-PCR. One experiment representative of three experiments. (C) The protein expression of IL-17RA, IL-17RC, and ACT1 in cultured RPE cells were detected by ﬂow cytometry. One experiment representative of three experiments. (D) Immunoﬂuorescence analysis of IL-17RA and IL-17RC in human retinal tissue. Retinal sections from healthy human donor eyes were stained with either goat polyclonal anti–IL-17RA or rabbit monoclonal anti–IL-17RC as indicated. Negative controls were

Downloaded from tvst.arvojournals.org on 10/01/2021 IL-17A Induces IL-1b Secretion From RPE Cells IOVS j February 2016 j Vol. 57 j No. 2 j 315

generated by omission of the primary antibody. Secondary antibodies were coupled to Alexa Fluor 488 (green) and Alexa Fluor 555 (red) respectively. Nuclei were stained with Hoechst 33342. Sections were analyzed under a confocal microscope. BrM, Bruch’s membrane; INL, inner nuclear layer; ONL, outer nuclear layer; Ch, choroid.

turer’s instructions. A total of 1 lg total RNA isolated was Flow Cytometry reverse transcribed with the reverse transcriptase kit (Takara Bio, Inc., Shiga, Japan) according to the manufacturer’s To detect IL-17RA and IL-17RC expression, RPE cells were protocol. IL-17RA, IL-17RC, ACT1, NLRP3, and GAPDH were stained with anti–IL-17RA or anti–IL-17RC, and then were amplified using ExTaq DNA polymerase (Takara Bio, Inc.). stained with FITC-conjugated anti-mouse IgG. To analyze the Polymerase chain reaction assay cycles were as follows: 948C expression of ACT1, RPE cells were fixed and permeabilized in Intracellular Fixation & Permeabilization Buffer Set (eBio- for 5 minutes, 35 cycles of 948C for 30 seconds, 568C for 30 science). Cells then were stained with anti-ACT1 and FITC- seconds, and 728C for 30 seconds. The PCR products were conjugated anti-mouse IgG. For the detection of phospho-NF-jB visualized on 2% agarose gels and ethidium bromide staining. p65, the RPE cells were stimulated with 100 ng/mL IL-17A for 5, To check the mRNA levels of pro-IL-1b, IL-8, CCL2, CCL20, 15, or 30 minutes. The cells then were cooled quickly, fixed and , and in RPE cells, real-time RT-PCR was pro-caspase-1 NLRP3 permeabilized with BD Cytofix/Cytoperm solution (BD Biosci- performed in triplicate with a real-time RT-PCR system (ABI ences, Franklin Lakes, NJ, USA), then sequentially stained with PRISM 7500; Applied Biosystems, Foster City, CA, USA) using a PE-labeled phospho-NF-jB p65 antibody (BD Biosciences). An SYBR detection kit (Takara Bio, Inc.) according to the standard isotype control mouse IgG was used for gating of phospho- protocol. Specific primers for NLRP3 and GAPDH were protein positive cells. For ROS measurement, treated RPE cells identical to those used for RT-PCR. The mRNA levels of each were incubated with 10 mM CM-H2DCFDA (Invitrogen) for 30 target gene were normalized to the levels of GAPDH and were minutes at 378C, harvested, and analyzed. FACS data analysis was represented as fold induction. The primer sequences of for RT- performed using the Diva (BD Biosciences) and FlowJo PCR and real-time RT-PCR are shown in the Table. (TreeStar, Ashland, OR, USA) software.

Western Blot Enzyme-Linked Immunosorbent Assay (ELISA) After treatment, RPE cells were harvested and lysed with RIPA Retinal pigment epithelial cells were seeded in 12-well plates buffer (Shenggong, Shanghai, China). Cytosol proteins from and grown to 80% confluence. After treatment, supernatants RPE cells (40 lg/lane) were separated by electrophoresis in 8% were collected and subsequently centrifuged to pellet cell to 12% SDS-polyacrylamide gel and then transferred onto a debris. Supernatants were analyzed using human ELISA polyvinylidine fluoride (PVDF) membrane (Millipore, Bedford, development kits for IL-1b (R&D Systems) according to the MA, USA). Subsequently, the membrane was blocked in 2% BSA manufacturer’s instructions. for 1 hour at room temperature and incubated with the specific primary antibodies (diluted 1:100–1:1000) overnight at 48C. KnockDown of NLRP3 Expression After washing with phosphate buffer solution containing Tween-20 (PBST) five times, the membranes were probed To reduce endogenous NLRP3 expression, RPE cells were with horseradish peroxidase-conjugated secondary antibodies transfected with 80 pmol small interfering RNA (siRNA) (diluted 1:5000 by 2% BSA). The level of b-actin also was oligonucleotides (Santa Cruz Biotechnology) following the measured at the same time as an internal control. The results manufacturer’s instructions. The cells were transfected with were quantified by analysis for gray scale using Gel-Pro either a siRNA oligonucleotide against NLRP3 or a nontargeted Analyzer software (Media Cybernetics, Inc., Rockville, MD, control siRNA oligonucleotide. After incubation, the RPE cells USA). were stimulated with IL-17A (100 ng/mL) for 24 hours, the cell- culture medium was collected for IL-1b ELISA analysis, and the cells were harvested for the detection of caspase-1 and NLRP3 expression using RT-PCR and Western blot.

Statistical Analysis Results were expressed as the mean 6 SEM. A 1-way ANOVA was used to compare variances within and among groups. Data were evaluated statistically using post hoc 2-tailed Student’s t- tests. Statistical signiﬁcance was set at P < 0.05.

FIGURE 2. IL-17A induces the transcription of pro-IL-1b and secretion RESULTS of IL-1b in RPE cells. (A) Retinal pigment epithelial cells were stimulated by 100 ng/mL IL-17A for 24 hours. Pro-IL-1b, IL-8, CCL2, IL-17A Receptors are Constitutively Expressed on and CCL20 gene expressions then were assessed by real-time RT-PCR and normalized against the amount of GAPDH mRNA. Gene expression RPE Cells is graphed as mean fold induction over medium control. The data IL-17A binds to a heterodimeric receptor consisting of IL-17RA represent the mean 6 SEM of three experiments with similar results and IL-17RC, and the downstream signaling pathway requires (*P < 0.05, **P < 0.01). (B) Retinal pigment epithelial cells were stimulated by different concentrations of IL-17A for 24 hours. Mature an essential signaling adaptor, ACT1 (also known as TNFR- 25 IL-1b secretion into culture supernatant was assessed using ELISA. The associated factor [TRAF]3IP2 or CIKS). To determine data represent the mean 6 SEM of three experiments with similar whether human RPE cells might be responsive to IL-17A, we results (*P < 0.05, **P < 0.01). isolated and cultured primary human RPE cells (Fig. 1A), and

Downloaded from tvst.arvojournals.org on 10/01/2021 IL-17A Induces IL-1b Secretion From RPE Cells IOVS j February 2016 j Vol. 57 j No. 2 j 316

FIGURE 3. The role of NF-jB signaling in the upregulation of pro-IL-1b mRNA in RPE cells. (A) Kinetic of Akt, Erk1/2, p38 MAPK, and NF-jB p65 phosphorylations was determined by Western blot in RPE cells treated by 100 ng/mL IL-17A. One experiment representative of three experiments. (B) Retinal pigment epithelial cells were exposed to 100 ng/mL IL-17A for indicated time. NF-jB p65 phosphorylation was determined by ﬂow cytometry. Values denote phospho-p65þ cells (%) from three independent experiments. (C) Retinal pigment epithelial cells were treated with IL-17A (100 ng/mL), in the presence NF-jB inhibitor Bay 11-7082 in indicated concentrations (5 and 10 lM). After 24 hours, the cells were analyzed for pro- IL-1b mRNA levels by real-time RT-PCR. Gene expression is graphed as mean fold induction over medium control. The data represent the mean 6 SEM of three experiments with similar results (*P < 0.05, **P < 0.01).

then performed RT-PCR to detect mRNA transcripts of IL- production of IL-1b involves the induction of pro-IL-1b and the 17RA, IL-17RC, and ACT1. As shown in Figure 1B, RPE cells processing of pro-IL-1b into mature IL-1b. In most cases, the constitutively expressed IL-17RA, IL-17RC, and ACT1 mRNA. induction of pro-IL-1b requires the activation of NF-jB Furthermore, analyzing their protein expression by flow signaling cascade. Next, we sought to determine whether NF- cytometry confirmed that RPE cells constitutively expressed jB could be responsible for IL-17A–induced pro-IL-1b upreg- IL-17RA, IL-17RC, and ACT1 (Fig. 1C). To rule out the ulation in RPE cells. Using flow cytometry, we confirmed that possibility that the expression of these receptors in RPE cells NF-jB p65 phosphorylation was enhanced in IL-17A–treated was due to culture conditions, we performed immunofluores- RPE cells (Fig. 3B). Furthermore, using Bay 11-7082, an NF-jB cence using healthy retinal tissue. A positive staining of IL- inhibitor, we demonstrated that IL-17A–induced pro-IL-1b 17RA and IL-17RC was observed in RPE layer (Fig. 1D). These mRNA was significantly suppressed (Fig. 3C). These results results suggest that RPE cells may be directly influenced by IL- suggest that activation of NF-jB signaling is involved in the IL- 17A via signaling through heterodimeric IL-17RA and IL-17RC. 17A–mediated of pro-IL-1b induction.

IL-17A Triggers IL-1b Secretion From RPE Cells IL-17A Activates the NLRP3 Inflammasome in RPE We postulated that IL-17A might be an inducer of inflammatory Cells mediators in human RPE cells. To verify this assumption, RPE cells were incubated with IL-17A (100 ng/mL) for 24 hours, Inflammasome activation results in the processing of pro-IL-1b and the mRNA levels of pro-IL-1b, IL-8, CCL2, and CCL20 were into the mature IL-1b on the posttranscriptional level. Next, we determined by real-time RT-PCR. IL-17A significantly upregu- aimed to study NLRP3 inflammasome complex involved in IL- lated the mRNA expression of pro-IL-1b, IL-8, CCL2, and 17A–triggered IL-1b production. Retinal pigment epithelial CCL20 (Fig. 2A). IL-1b has been shown to be one of the key cells upregulated the expression of pro-caspase-1 and NLRP3 cytokines activating Th17 cells. In addition, the production of mRNA in response to IL-17A (Fig. 4A). Interleukin-17A also mature IL-1b from RPE cells stimulated by IL-17A was assayed induced higher amounts of active caspase-1 subunit (p20) and by ELISA. As shown in Figure 2B, IL-17A had a stimulatory NLRP3 at protein levels, suggesting NLRP3 inflammasome effect on the production of mature IL-1b in a dose-dependent activation (Fig. 4B). Interleukin-1b secretion following IL-17A manner. treatment was inhibited in a concentration-dependent manner by Z-YVAD-fmk, the specific inhibitor of caspase-1, confirming IL-17A Upregulates Pro-IL-1b mRNA in RPE Cells the role of caspase-1 in IL-17A–induced IL-1b secretion (Fig. 4C). To further examine the role of NLRP3 in IL-17A–triggered j Via NF- B IL-1b release from RPE cells, we adopted siRNA. Transfection To further explore the responsiveness of RPE cells to IL-17A, with siRNAs efficiently downregulated NLRP3 mRNA and we analyzed the downstream signaling molecules activated protein (Fig. 4D) expression in IL-17A–treated RPE cells. using Western blot. As shown in Figure 3A, significant Knockdown of NLRP3 significantly inhibited IL-17A-mediated phosphorylation of Akt, Erk1/2, p38 MAPK, and NF-jB p65 IL-1b release, indicating a crucial role of the NLRP3 inflamma- was detected in RPE cells following treatment of IL-17A. The some in the response of RPE cells to IL-17A (Fig. 4E).

Downloaded from tvst.arvojournals.org on 10/01/2021 IL-17A Induces IL-1b Secretion From RPE Cells IOVS j February 2016 j Vol. 57 j No. 2 j 317

DISCUSSION In this study, we determined that primary human RPE cells constitutively expressed IL-17RA, IL-17RC, and ACT1. Treatment with IL-17A caused an increase in IL-1b production via ROS production leading to NLRP3 activation in RPE cells. Recent efforts have drawn attention to the Th17 cells and Th17 cytokines, particularly IL-17A, as a pathogenic factor in AMD. Serum levels of IL-17A were reported to be significantly higher in AMD patients compared to normal controls, while serum levels of IL-22, another Th17 family cytokine, also were significantly higher in AMD patients. The elevation of IL-17A expression was found not only in the peripheral blood but also in macular tissues with AMD lesions.17–19 Since AMD patients usually present without systemic inflammation, we hypothesized that local IL- 17A signaling was more relevant in AMD pathogenesis. Although Th17 cells, a subtype of activated CD4þ T cells, often are deemed the major producers of IL-17A, the source of IL-17A in the outer retina is not known. Other immune competent cells, such as cdT cells, neutrophils, and even macrophages, under specific conditions may produce IL- 17A.28–30 In a recent study, using Nrf2/ mice as a model FIGURE 4. The involvement of NLRP3 inflammasome in the produc- 31 tion of IL-1b by RPE cells stimulated with IL-17A. (A) Retinal pigment with pathologic features similar to human AMD, Zhao et al. epithelial cells were stimulated with 100 ng/mL IL-17A for 24 hours, showed that lymphocytes infiltrated in the sub-RPE space were and the mRNA expression of pro-caspase-1 and NLRP3 was mainly cdT cells, which produced IL-17A as a proinflammatory determined using real-time RT-PCR and normalized against the amount signal. Histologic evidence has shown the presence of of GAPDH mRNA. Gene expression is graphed as mean fold induction macrophages near many AMD lesions. Therefore, it is over medium control. The data represent the mean 6 SEM of three worthwhile to determine further the exact cellular sources of independent experiments (**P < 0.01). (B) Retinal pigment epithelial local IL-17A in the progression of AMD.32 cells were treated with 100 ng/mL IL-17A for 24 hours. The cells were The elevated expression of IL-17RC recently was observed lysed and the whole-cell extracts were assessed by Western blot using 17 antibodies against NLRP3 and the active form of caspase-1 (p20). in chorioretinal tissues from AMD patients. Therefore, the Representative blots of three independent experiments are shown. (C) RPE cells, being resident at the chorioretinal interface, would Retinal pigment epithelial cells were treated with caspase-1 inhibitor Z- appear a potential target of IL-17A immunologic effect. A YVAD-fmk in indicated concentrations (1 and 10 lg/mL) for 30 minutes recent report showed that IL-17A promoted ARPE-19 cells to and then stimulated with 100 ng/mL IL-17A. Culture supernatants were secrete inflammatory mediators and compromised the ARPE-19 collected after 24 hours and assayed with IL-1b ELISA. The data monolayer barrier function.20 In this report, Chen el al.20 found represent the mean 6 SEM of three experiments with similar results ARPE-19 cells express IL-17RC but no IL-17RA, while in our (*P < 0.05, **P < 0.01). (D) Retinal pigment epithelial cells were study, we demonstrated that RPE cells express IL-17RA and IL- transfected with siRNA oligos specific for NLRP3 (siNLRP3) or a nonspecific siRNA oligo (siControl), and subsequently stimulated with 17RC, which are essential to IL-17A signaling. This discrepancy 100 ng/mL IL-17A for 24 hours, and the levels of NLRP3 mRNA (above) could be explained by the origin of the RPE cells adopted in and protein (below) were analyzed using RT-PCR and Western blot. the studies. ACT1 has been implicated in the control of One experiment representative of three experiments. (E)Small immune and inflammatory responses. The binding of IL-17A to interfering RNA–transfected RPE cells then were stimulated with 100 the IL-17RA/RC heterodimer leads to the ACT1 recruitment ng/mL IL-17A, and then culture supernatants were collected after 24 through a SEFIR-SEFIR-dependent interaction. This results in hours and assayed with IL-1b ELISA. The data represent the mean 6 incorporation of TRAF6 into the signaling complex and the SEM of three experiments with similar results (*P < 0.05). subsequent downstream activation of the NF-jB and MAPK pathways, leading to the induction of target genes.24 The expression of ACT1 in RPE cells has never been characterized before to our knowledge, and our results provided support for ROS Generation is Required for IL-17A-Induced IL- the RPE cells to be targets of IL-17A. 1b Secretion From PRE Cells The most notable effects of IL-17A on RPE cells might be the increased production of inflammatory cytokines and chemo- Reactive oxygen species (ROS) have been proposed to perform kines. In current study, IL-17A was able to significantly stimulate an important role in the activation of NLRP3 inflammasome, the expression of IL-8, CCL2,andCCL20 by RPE cells. Our and IL-17A has been reported to be an inducer of ROS.26,27 results were consistent with previous studies showing the Therefore, we hypothesized that ROS formation in RPE cells enhanced production of these cytokines by IL-17A.20 These might be involved in IL-17A–mediated inflammasome activa- cytokines and chemokines have been implicated previously in tion and IL-1b secretion. Indeed, as determined by CM- the development of AMD. A strong correlation has been H2DCFDA labeling and flow cytometry, ROS formation in obtained between elevated levels of IL-8 concentrations in aqueous humor and exudative AMD patients.33 Murine models RPE cells increased significantly following stimulation with IL- have shown that CCL2 is involved in drusen formation and RPE 17A (Fig. 5A). Preconditioning with the antioxidant NAC changes seen in the early stages of AMD.34 CCL20 is a significantly modulated the expression of caspase-1 and NLRP3 chemokine involved in the attraction of Th17 cells. These at mRNA and protein levels (Figs. 5B, 5C). Furthermore, IL- cytokines and chemokines might promote immune cells 17A–induced IL-1b production was significantly inhibited by trafficking into the inflammatory sites within macular tissues NAC. These data indicated that activation of the NLRP3 of AMD patients and eventually causing the pathology of AMD. inflammasome by IL-17A is dependent on ROS generation in The activation of MAPK and Akt pathways might be an RPE cells. important signaling event in the response of RPE cells to

Downloaded from tvst.arvojournals.org on 10/01/2021 IL-17A Induces IL-1b Secretion From RPE Cells IOVS j February 2016 j Vol. 57 j No. 2 j 318

FIGURE 5. IL-17A activation of inflammasome and IL-1b secretion in RPE cells is regulated by ROS. (A) Retinal pigment epithelial cells were incubated with 100 ng/mL IL-17A for 6 and 24 hours. The cells then were collected and labeled with CM-H2DCFDA, and ROS production was measured by flow cytometry. Representative histograms and mean fluorescence intensities (mean 6 SEM) were obtained from five independent experiments (*P < 0.05, **P < 0.01). (B) Retinal pigment epithelial cells were treated with 100 ng/mL IL-17A in combination of NAC (5 mM) for 24 hours, and the mRNA levels of pro-caspase-1 and NLRP3 were determined using real-time RT-PCR and normalized against the amount of GAPDH mRNA. Gene expression is graphed as mean fold induction over medium control. The data represent the mean 6 SEM of three experiments with similar results (*P < 0.05). (C) Retinal pigment epithelial cells were treated with 100 ng/mL IL-17A for 24 hours, in the presence of NAC (5 mM). NLRP3 and the active form of caspase-1 (p20) then were analyzed by Western blot. Representative blots of three independent experiments are shown. (D) Retinal pigment epithelial cells were treated with 100 ng/mL IL-17A in the presence of 5 mM NAC, and then culture supernatants were collected after 24 hours and assayed with IL-1b ELISA. The data represent the mean 6 SEM of three experiments with similar results (*P < 0.05).

proinflammatory cytokines.35 In the present study, activation of The induction of NLRP3 expression via activation of NF-jB Erk1/2, p38 MAPK and Akt pathways was observed in IL-17A– is a critical checkpoint for NLRP3 inflammasome activation.4 treated RPE cells, which are indicated by elevated phosphory- Accordingly, in the present study, we observed the upregulated/total protein ratio. Our findings are in accordance with lation of NLRP3 and p65 expression following the treatment of earlier reports using ARPE-19 cell line.35 The promoter region of IL-17A, highlighting an IL-17A–induced activation of NLRP3 the human pro-IL-1b gene has been cloned and has been shown inflammasome. The activation of NLRP3 inflammasome re- to contain consensus binding motifs for NF-jB.36 Here, we quires two signals, a ‘‘priming signal’’ and an ‘‘activation showed that Bay 11-7082 reduced pro-IL-1b expression in RPE signal.’’ The ‘‘priming signal’’ majorly channels through NF-jB, cells, suggesting a crucial role for NF-jB signaling as the initial upregulating the transcription of pro-IL-1b. The second signal step in the NLRP3 inflammasome activation. for inflammasome activation is relayed by various mechanisms IL-1b performs a fundamental role in orchestrating inflam- including generation of intracellular ROS. Retinal pigment matory responses, majorly via the regulation of gene expres- epithelial cells are major targets of oxidative stress because of sion. IL-1b–responsive genes coordinate all phases of local their high oxygen consumption, high levels of polyunsaturated inflammation and also attract and activate cells of the adaptive lipids, and the long periods of exposure to light. Disturbances immune system. Th17 cells possess IL-1b receptors and are in the oxidant-antioxidant system in retina may have an stimulated by IL-1b to generate IL-17A. Therefore, excess IL-1b important role in the pathogenesis of AMD.37 Here, we production might result in the augmentation of Th17-dominant demonstrated that ROS generation was necessary for NLRP3 pathology. Myloid cells serve as major producers of this Th17- inflammasome activation in RPE cells and subsequent IL-1b promoting cytokine. However, our data offered convincing release in response to IL-17A. Therefore, we hypothesize that support for RPE cells as an important source of IL-1b, especially IL-17A activates RPE cells in a positive feedback loop in the early development of AMD with marginal infiltration of producing inflammatory mediators and ROS. One of the major myloid cells in the retinal tissue. To our knowledge, no studies limitations of present study is lack of in vivo comparison of IL- to date have adequately addressed the possibility of a feedback 17A as well as NLRP3 inflammasome components between mechanism occurring via the activity of IL-17A on IL-1b healthy and AMD retinal tissue. More in vivo evidences will be secreted from RPE cells under conditions of retinal inflamma- needed to confirm the hypothesis in the context of AMD, and tion. to decipher how the local microenvironment and RPE-derived Mature IL-1b is generated via the cleavage of the inactive pro- cytokines influence IL-17A-producing cells. IL-1b precursor by caspase-1 (p20). The inflammasome complex- In summary, the present study revealed the presence of IL- es activate caspase-1 by facilitating the cleavage of its zymogen 17RA, IL-17RC, and ACT1 in cultured human RPE cells. precursor, pro-caspase-1. Active caspase-1 (p20) then catalyzes Interleukin-17A stimulated RPE cells to produce mature IL-1b the proteolytic cleavage of pro-IL-1b into IL-1b.4 The NLRP3 through activation of NF-jB pathway and NLRP3 inflamma- inflammasome-mediated activation of caspase-1 and processing of some, thereby supporting a mechanism bridging between pro-IL-1b into IL-1b has been reported previously in RPE cells.14 innate and adaptive immune responses in AMD. This mecha- In our study, Z-YVAD-fmk abolished the caspase-1 activity and nism must be addressed in future investigations, but this study reduced the IL-1b production, suggesting that the IL-17A-induced may contribute to our understanding of retinal immunity and IL-1b release from RPE cells also was mediated by caspase-1. translate into better therapy for AMD.

Downloaded from tvst.arvojournals.org on 10/01/2021 IL-17A Induces IL-1b Secretion From RPE Cells IOVS j February 2016 j Vol. 57 j No. 2 j 319

Acknowledgments its implication in age-related macular degeneration. J Transl Med. 2011;9:1–12. Supported by the National Natural Science Foundations of China 19. Wei L, Liu B, Tuo J, et al. Hypomethylation of the IL17RC (81200675, 81470624) and the Fundamental Research Funds for promoter associates with age-related macular degeneration. Cell the Central Universities. The authors alone are responsible for the Rep. 2012;2:1151–1158. content and writing of the paper. 20. Chen Y, Yang P, Li F, Kijlstra A. The effects of Th17 cytokines on Disclosure: S. Zhang, None; N. Yu, None; R. Zhang, None; S. the inflammatory mediator production and barrier function of Zhang, None; J. Wu, None ARPE-19 cells. PLoS One. 2011;6:e18139. 21. Leung KW, Barnstable CJ, Tombran-Tink J. Bacterial endotoxin References activates retinal pigment epithelial cells and induces their degeneration through IL-6 and IL-8 autocrine signaling. Mol 1. Nowak JZ. AMD-the retinal disease with an unprecised etio- Immunol. 2009;46:1374–1386. pathogenesis: in search of effective therapeutics. Acta Pol 22. Sutton CE, Lalor SJ, Sweeney CM, Brereton CF, Lavelle EC, Mills Pharm. 2014;71:900–916. KH. Interleukin-1 and IL-23 induce innate IL-17 production from 2. Camelo S. Potential sources and roles of adaptive immunity in gammadelta T cells, amplifying Th17 responses and autoimmu- age-related macular degeneration: shall we rename AMD into nity. Immunity. 2009;31:331–341. autoimmune macular disease? Autoimmune Dis. 2014;2014: 23. Zielinski CE, Mele F, Aschenbrenner D, et al. Pathogen-induced 532487. human TH17 cells produce IFN-c or IL-10 and are regulated by IL- 3. Nussenblatt RB, Lee RW, Chew E, et al. Immune responses in age- 1b. Nature. 2012;484:514–518. related macular degeneration and a possible long-term therapeu- 24. Maminishkis A, Chen S, Jalickee S, et al. Confluent monolayers of tic strategy for prevention. Am J Ophthalmol. 2014;158:5–11. cultured human fetal retinal pigment epithelium exhibit mor- 4. Davis BK, Wen H, Ting JP. The inflammasome NLRs in immunity, phology and physiology of native tissue. Invest Ophthalmol Vis inflammation, and associated diseases. Ann Rev Immunol. 2011; Sci. 2006;47:3612–3624. 29707–29735. 25. Gu C, Wu L, Li X. IL-17 family: cytokines, receptors and signaling. 5. Marneros AG. NLRP3 inflammasome blockade inhibits VEGF-A- Cytokine. 2013;64:477–485. induced age-related macular degeneration. Cell Rep. 2013;4:945– 958. 26. Martinon F. Signaling by ROS drives inflammasome activation. Eur J Immunol. 2010;40:616–619. 6. Kerur N, Hirano Y, Tarallo V, et al. TLR-independent and P2X7- dependent signaling mediate Alu RNA-induced NLRP3 inflamma- 27. Cho KA, Suh JW, Lee KH, Kang JL, Woo SY. IL-17 and IL-22 some activation in geographic atrophy. Invest Ophthalmol Vis enhance skin inflammation by stimulating the secretion of IL-1b Sci. 2013;54:7395–7401. by keratinocytes via the ROS-NLRP3-caspase-1 pathway. Int . 2012;24:147–158. 7. Hu SJ, Calippe B, Lavalette S, et al. Upregulation of P2RX7 in Immunol Cx3cr1-deficient mononuclear phagocytes leads to increased 28. Hamada S, Umemura M, Shiono T, et al. IL-17A produced by interleukin-1b secretion and photoreceptor neurodegeneration. J gammadelta T cells plays a critical role in innate immunity Neurosci. 2015;35:6987–6996. against listeria monocytogenes infection in the liver. J Immunol. 8. Anderson OA, Finkelstein A, Shima DT. A2E induces IL-1b 2008;181:3456–3463. production in retinal pigment epithelial cells via the NLRP3 29. Werner JL, Gessner MA, Lilly LM, et al. Neutrophils produce inflammasome. PLoS One. 2013;8:e67263. interleukin 17A (IL-17A) in a dectin-1- and IL-23-dependent 9. Holtkamp GM, Kijlstra A, Peek R, de Vos AF. Retinal pigment manner during invasive fungal infection. Infect Immun. 2011;79: epithelium-immune system interactions: cytokine production 3966–3977. and cytokine-induced changes. Prog Retin Eye Res. 2001;20:29– 30. Song C, Luo L, Lei Z, et al. IL-17-producing alveolar macrophages 48. mediate allergic lung inflammation related to asthma. J 10. Kaarniranta K, Salminen A. Age-related macular degeneration: Immunol. 2008;181:6117–6124. activation of innate immunity system via pattern recognition 31. Zhao Z, Xu P, Jie Z, et al. cd T cells as a major source of IL-17 receptors. J Mol Med (Berl). 2009;87:117–123. production during age-dependent RPE degeneration. Invest 11. Halle A, Hornung V, Petzold GC, et al. The NALP3 inflammasome Ophthalmol Vis Sci. 2014;55:6580–6589. is involved in the innate immune response to amyloid-beta. Nat 32. Killingsworth MC, Sarks JP, Sarks SH. Macrophages related to Immunol. 2008;9:857–865. Bruch’s membrane in age-related macular degeneration. Eye 12. Sutterwala FS, Haasken S, Cassel SL. Mechanism of NLRP3 (Lond). 1990;4:613–621. inflammasome activation. Ann N Y Acad Sci. 2014;1319:82–95. 33. Sakurada Y, Nakamura Y, Yoneyama S, et al. Aqueous humor 13. Tarallo V, Hirano Y, Gelfand BD, et al. DICER1 loss and Alu RNA cytokine levels in patients with polypoidal choroidal vasculop- induce age-related macular degeneration via the NLRP3 inflam- athy and neovascular age-related macular degeneration. Oph- masome and MyD88. Cell. 2012;149:847–859. thalmic Res. 2015;53:2–7. 14. Tseng WA, Thein T, Kinnunen K, et al. NLRP3 inflammasome 34. Falk MK, Singh A, Faber C, Nissen MH, Hviid T, Sørensen TL. activation in retinal pigment epithelial cells by lysosomal CX3CL1/CX3CR1 and CCL2/CCR2 chemokine/chemokine re- destabilization: implications for age-related macular degenera- ceptor complex in patients with AMD. PLoS One. 2014;9: tion. Invest Ophthalmol Vis Sci. 2013;54:110–120. e112473. 15. Kauppinen A, Niskanen H, Suuronen T, Kinnunen K, Salminen A, 35. Chen Y, Kijlstra A, Chen Y, Yang P. IL-17A stimulates the Kaarniranta K. Oxidative stress activates NLRP3 inflammasomes production of inflammatory mediators via Erk1/2, p38 MAPK, in ARPE-19 cells-implications for age-related macular degenera- PI3K/Akt, and NF-jB pathways in ARPE-19 cells. Mol Vis. 2011; tion (AMD). Immunol Lett. 2012;147:29–33. 17:3072–3077. 16. Miossec P, Korn T, Kuchroo VK. Interleukin-17 and type 17 36. Hiscott J, Marois J, Garoufalis J, et al. Characterization of a helper T cells. N Engl J Med. 2009;361:888–898. functional NF-kappa B site in the human interleukin 1 beta 17. Ardeljan D, Wang Y, Park S, et al. Interleukin-17 retinotoxicity is promoter: evidence for a positive autoregulatory loop. Mol Cell prevented by gene transfer of a soluble interleukin-17 receptor Biol. 1993;13:6231–6240. acting as a cytokine blocker: implications for age-related macular 37. Tokarz P, Kaarniranta K, Blasiak J. Role of antioxidant enzymes degeneration. PLoS One. 2014;9:e95900. and small molecular weight antioxidants in the pathogenesis of 18. Liu B, Wei L, Meyerle C, et al. Complement component C5a age-related macular degeneration (AMD). Biogerontology. 2013; promotes expression of IL-22 and IL-17 from human T cells and 14:461–82.

Downloaded from tvst.arvojournals.org on 10/01/2021