Upregulation of the IL-33/ST2 pathway in dry eye

IL-33, a newly IL-1 family cytokine

In recent years, researchers have found Th2 CD4⁺ T cells also have an association with dry eye, and have a role in maintaining homeostatic levels of conjunctival goblet cell [2]. One cytokine that stimulates Th2 cell is IL-33, which is a newly discovered cytokine that belongs to the IL-1 family [8]. It has been reported that IL-33 is involved in the pathophysiology of allergic diseases, cardiovascular disease, rheumatoid arthritis, and chronic obstructive pulmonary disease, and plays different roles in these inflammatory diseases [9].

IL-33 can be released through cell death or stress, inducing excessive inflammatory responses. Recent studies have clarified the distinct function of IL-33 as a nuclear factor, in addition to being an extracellular cytokine [10]. IL-33 is consistently expressed in the nucleus of endothelial cells, epithelial cells, fibroblasts, and macrophages [8,10]. Full length IL-33 is bioactive [11], but proteases derived from different cellular sources, such as neutrophils, macrophages, or mast cells, process bioactive full-length IL-33 into N-terminally truncated forms (hIL-33_95–270, hIL-33_99–270, and hIL-33_109–270) that have up to 30-fold higher biologic activity than full-length IL-33. IL-33 is initiated by binding with ST2 [8], a member of the IL-1 receptor family, which is expressed (or induced) on various immune cell types, including mast cells, basophils, eosinophils, neutrophils, macrophages, dendritic cells, Th2 lymphocytes, invariant natural killer T (iNKT), and natural killer (NK) cells [12-14].

ST2 has two different isoforms

ST2L and sST2 are generated by alternative splicing of ST2 mRNA. ST2L is a membrane-bound form that activates the intracellular signaling pathways, while sST2 is a soluble form that inhibits IL-33 signaling by acting as a decoy receptor. After binding with ST2, IL-33 leads to recruitment of MyD88, interleukin-1 receptor-associated kinase (IRAK), and TNF receptor-associated factor 6 (TRAF6), and sequentially activates downstream MAP kinase (MAPK), c-Jun N-terminal kinase (JNK), p38, extracellular signal-regulated kinase (ERK), and NF-κB signaling pathways, and then induces cytokine production in target cells [15]. As Th2 cells are a major target of IL-33, IL-33 binds to the ST2L of Th2 cells to stimulate the secretion of the Th2 cytokines, such as IL4, IL-5, and IL-13 [16].

Because of the diffuse distribution of IL-33 and ST2 in many tissues and organs, the role of IL-33/ST2 signaling has been investigated in various inflammation-related diseases. However, little is known about IL-33 in the pathogenesis of DE. In this study, we demonstrated the robust expression of IL-33 in the tears of patients with DE, which might contribute to the progression of DE.

Patients

Twenty-five patients with dry eye (18 women and 7 men; mean age: 44.25 ± 10.22) and twenty healthy individuals (14 women and 7 men; mean age: 48.52 ± 10.03) matched for sex and age were enrolled for testing studies between March 2018 and September 2018 in the First Affiliated Hospital of Harbin Medical University. All subjects are in fine physical condition during the recruitment period. Written informed consent was obtained from all subjects before participation.

All subjects were informed of the aims of the study, and their written informed consent was obtained according to the Declaration of Helsinki. The human-related activities were adherent to the ARVO for the Use of Humans in Ophthalmic and Vision Research. We recruited Chinese subjects aged between 18 and 65 years. Dry eye was diagnosed according to criteria reported previously [17]: patients with (1) an ocular surface disease index (OSDI) score≥13, (2) tear film breakup time (TBUT)≤5 s or Schirmer I test≤5 mm, or (3) positive cornea fluorescein staining (CFS) [18]. The following exclusion criteria was used: (1) any intraocular inflammation condition, ocular surgery or trauma within the previous 6 months; (2) present ocular infection or allergy; (3) laser treatment within the previous 6 months; (4) ocular therapies other than artificial tears; (5) contact lens wear; (6) abnormalities in the cornea, conjunctiva, or eyelid; (7) any systemic diseases, such as cancer, AIDS, or hepatic disease.

Tear film breakup time, CFS, and Shirmer I test

TBUT was measured using prepackaged, sterile fluorescein paper strips (Jingming New Technological Development Co. Ltd, Tianjin, China). The strip was wetted with 20 μl saline, and then instilled into the tear film. The patients were asked to blink several times to ensure the dye was well mixed. The time from the last complete blink to the appearance of the first corneal black spot in the stained-tear film was measured three times, and the mean value of the measurements was calculated. After the TBUT was measured, the CFS score was evaluated. CFS scores higher than 3 points were regarded as abnormal. The Schirmer I test was performed without anesthesia. The rounded bulb end of the strip was folded and placed in the lateral canthus, away from the cornea. After wetting for 5 min, the length of the wet filter paper was recorded. The strip was placed in a 0.5-ml tube to be stored immediately at −80 °C.

Tear collection

According to the method described previously [19], tears were extracted from the Schirmer strips as follows: 100 μl PBS (1X; 155 mM NaCl, 3 mM NaHPO₄, 1 mM KHPO₄, pH 7.4) was added to the strips, and allowed to solubilize for 1 h at 4 °C. Tears were obtained by puncturing the tube at the bottom with a sterilized needle. Then the tube was put into a 1.5-ml tube, and centrifuged at 10,000 ×g at 4 °C for 10 min to obtain the tears.

Conjunctival impression cytology

Conjunctival impression cytology (CIC) samples were collected according to the method described previously [20,21]. The CIC specimens were obtained after administration of topical anesthesia with 0.4% oxybuprocaine. Two separate sterile membrane filters (0.45 μm; Millipore, Boston, MA) were applied to adjacent superior and inferior temporal bulbar conjunctiva, pressed for 10 s gently with a forceps, and then removed. The filter paper collected from both eyes was transferred immediately to an empty 1.5-ml Eppendorf tube for western blotting, to evaluate the protein expression. All samples were immediately placed on ice until they were transferred to −80 °C for storage. The CIC samples were lysed with radioimmunoprecipitation assay buffer (1% Triton X-100, 1% deoxycholate, 0.1% sodium dodecyl sulfate [SDS]) on ice for 1 h, and then centrifuged at 10,000 ×g at 4 °C for 10 min to obtain the supernatant.

Primary culture and cell treatment of human conjunctival epithelial cells

Two pairs of donated normal human conjunctiva samples were preserved in Dulbecco's Modified Eagle Media (DMEM)/F12 (Thermo Fisher, Shanghai, China) before they were transferred to the laboratory. The tissues were washed three times in PBS with penicillin and streptomycin (5,000 U/ml), then cut into explants of 5 mm × 5 mm, and placed on six-well plates with the epithelium side down.

Approximately 15 min were allowed for attachment of the explants, and 100 μl of fetal bovine serum (FBS) was added to cover the explants overnight. The culture medium consisting of DMEM/F12, penicillin/streptomycin (5,000 U/ml), human epidermal growth factor (10 ng/ml), and 10% FBS was supplemented the next day. The primary cell cultures were incubated at 37 °C with 5% CO₂, and observed daily. The medium was replaced every 3 days. Sub-culturing was done when the cells reached 70–80% confluence. Sub-confluent cultures were switched to a serum-free medium (DMEM without FBS) for 24 h before treatment. Then the cells were cultured for an additional 4 or 24 h in an equal volume (1.0 ml/well) of serum-free medium with a different osmolarity, ranging from 400 to 500 mOsm, by adding 50, 70, and 90 mM sodium chloride (NaCl) [19]. Cells collected at 4 h were stored for real-time PCR. The IL-33 protein level was measured following 24 h incubation with western blotting.

Cell proliferation assay

To detect the effect of hyperosmolarity on cell activity, a cell proliferation assay kit (CCK-8; Dojindo, Shanghai, China) was used in accordance with the manufacturer’s instructions. Human conjunctival epithelial cells (HConECs) were seeded in 96-well plates (1 × 10⁴/well). A total of 10 μl CCK-8 solution was added after the cell treatment. Following a 4 h incubation, the plate was measured at 450 nm with a spectrophotometer (Bio-Rad Laboratories, Hercules, CA).

Western blotting

The protein level was determined using the BCA protein assay kit (Beyotime, Shanghai, China). After protein loading buffer was added and the samples were heated at 100 °C for 10 min, all samples were run on a 12% SDS-polyacrylamide gel electrophoresis (SDS–PAGE) gel, blotted onto a polyvinylidene fluoride (PVDF) membrane, and then blocked with 5% non-fat dried milk for 2 h at room temperature and incubated overnight at 4 °C with the following antibodies: anti-beta-actin (Abcam, Cambridge, MA, ab8226), anti-IL-33 antibody (Abcam, ab207737), anti-ST2 antibody (Santa Cruz Biotechnology, Dallas, TX, sc-74297). The membranes were then rinsed six times with Tris-buffered saline containing 0.5% Tween-20 (TBST), and incubated with goat anti-rabbit or anti-mouse immunoglobulin (IgG; Cell Signaling, Danvers, MA) secondary antibodies for 1 h at room temperature. The Odyssey fluorescent scanning system (LI-COR, Biorad) and Quantity One software were used to detect and analyze the immunoreactive proteins.

Quantitative real-time PCR

Total RNA was extracted using TRIzol reagent (Invitrogen, Shanghai, China) following the manufacturer’s instructions [22]. The isolated RNA was reverse-transcribed into synthesized cDNA in a 20 μl reaction mixture. Quantification of the IL-33 level with quantitative real-time PCR (RT-PCR) for 95°C, 2 min; 95°C,15s; 60°C, 15s; 72°C,45s; Number of cycles: 40 cycles was performed using the TaqMan miRNA assay system (Applied Biosystems, Foster City, CA). The primers were synthesized by Co Invitrogen Ltd. (Beijing, China). The sequences of primers are shown in Table 1. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was taken as a loading control. Data were analyzed using the 2^-△△Ct method.

Multiplexed analyses of cytokines in tears

The concentrations of the tear cytokines were determined measured with a Luminex MAGPIX system (Luminex Corporation, Chicago, IL) with the ProcartaPlex Human Assay kit (Thermo Fisher, Vienna, Austria). The system uses a liquid suspension array of four sets of beads. Each set of beads was combined with a monoclonal antibody raised against IL-33, IL-13, IL-5, and IL-4. The beads were prepared according to the manufacturer’s instructions. Standard curves were generated by using the reference cytokine concentrations supplied by the manufacturer. Results were analyzed with xPONENT software (Luminex Corporation) to obtain the concentration values.

Statistical analysis

Data were shown as mean ± standard deviation. Differences in group means were determined using one-way ANOVA. Statistical analyses were performed using SPSS (version 20.0; SPSS Inc., Chicago, IL), and GraphPad Prism 6 (GraphPad Software Inc., La Jolla, CA) was used to generate figures and tables. P values of less than 0.05 were considered statistically significant. Pearson correlation coefficients were used to assess correlations between changes in ocular surface parameters and tear cytokine levels in each group.

Validation of IL-33 in HConECs

To detect the response of HConECs to hyperosmolarity, CCK-8 was used to detect the cell viability. The mean of the cell viabilities was 86.401±6.3620, 72.517±11.382, and 69.930±5.5710 in 400, 450, and 500 mOsm, respectively (Figure 1A). The mean decreased with the increase in hyperosmolarity, but had no statistical significance. Real-time PCR was performed to analyze the mRNA expression of IL-33; western blotting was used to measure the protein level. According to the results, the expression of IL-33 mRNA seemed osmolality-dependent in the HConECs treated with hypertonic medium, and was 4.35-fold higher in the 500 mOsm medium compared with the 400 mOsm medium (p<0.001; One-way ANOVA; Figure 1B). Moreover, the production of the IL-33 protein was consistent with the transcription, and was 2.22-fold higher in the 500 mOsm medium compared with the 400 mOsm medium (p<0.01; One-way ANOVA; Figure 1C, D). The data shown are representative of the results in three independent experiments.

Demographics and clinical examination parameters

After discovering the phenomenon at the cellular level, we hoped to verify it on the human eye surface. A total of 45 subjects were enrolled in this study, including 25 patients with dry eye and 20 healthy controls. The ocular surface parameters are presented in Table 2. There were no statistically significant differences in the sex or mean age between the two groups. The mean OSDI score and CFS were statistically significantly higher in the DE group (p<0.0001; Student’s t-test), while the Schirmer test and the TBUT were much lower in the DE group than in the control group (p<0.0001; Student’s t-test).

IL-33/ST2 protein expression in conjunctival impression cytology

Because the IL-33/ST2 axis has a vital function in regulating inflammation and immune responses, we detected the expression in patients with dry eye. The samples were not pooled together, but were extracted and tested individually. With the use of western blotting, the relative expression of IL-33 (17 kDa) and ST2 (60 kDa) protein from CIC was higher in the DE group compared with the control group (IL-33: relative 9.25-fold upregulation, p<0.001; 17 and 15 samples in DE and controls respectively; ST2: relative 4.35-fold upregulation, p < 0.05, 15 and 19 samples in DE and controls respectively; Kruskal-Wallis test; Figure 2). All CIC results can be found in the Appendix 1.

Tear cytokine concentrations

The concentrations of IL-33, IL-13, IL-5, and IL-4 in the tears of each group were measured by Luminex. The mean levels of IL-33 in tears were 2.4850±0.6379 in controls, and 7.467±1.379 in the DE group (3.00-fold upregulation, p<0.0001; Figure 3A). The mean levels of IL-13 in tears were 0.2265±0.1223 in controls, and 1.5070±0.1048 in the DE group (6.65-fold upregulation, p<0.0001; Figure 3B). The mean levels of IL-5 in tears were 0.7905±0.3563 in controls, and 13.07±4.039 in the DE group (16.54 -fold upregulation, p=0.01; Figure 3C). As IL-33 can stimulate Th2 cells to release IL-13 and IL-5, we performed correlation analysis between IL-13, IL-5, and IL-33. The IL-13 and IL-5 levels were statistically significantly correlated with the IL-33 concentration (IL-33 with IL-5: R=0.9224, p<0.0001; IL-33 with IL-13: R=0.7263, p<0.0001; Figure 4). There were 21 and 17 samples in DE and control group respectively. We failed to detect the concentration of IL-4 as most samples were below the detection limit. Kruskal-Wallis test and Pearson correlation coefficients were used to analyze the data.

Correlations between cytokines and ocular surface parameters

The correlations between the levels of IL-33, IL-13, and IL-5 and ocular surface parameters, including the OSDI score, TBUT, Schirmer I test, and CFS, were evaluated in the DE group and healthy controls. In the DE group, the level of IL-33 was positively correlated with the OSDI score and CFS, but was negatively correlated with the Schirmer I test and the TBUT (IL-33 with OSDI: R=0.4483, p=0.0415; IL-33 with Shirmer I test: R=-0.4448, p=0.0433; IL-33 with the TBUT: R=-0.4459, p=0.0427; IL-33 with CFS: R=0.5657, p=0.0075; n=21; Figure 5A–D). The level of IL-13 was positively correlated only with CFS, and was negatively correlated with the Schirmer I test (IL-13 with Shirmer I test: R=-0.4348, p=0.0433; IL-13 with CFS: R=0.5005, p=0.0208; n=21; Figure 5E,F). The level of IL-5 was positively correlated with the OSDI score and CFS (IL-5 with OSDI: R=0.4551, p=0.0382; IL-5 with CFS: R=0.5261, p=0.0143; n=21; Figure 5G,H). IL-13 and IL-5 and the other ocular surface parameters showed no statistical significance. In the control group, the levels of IL-33, IL-13, and IL-5 in tears showed no correlation with the ocular surface parameters. Pearson correlation coefficients were used for analyzing,