For the present study, a total of 40 female C57BL/6 mice (8-week-old, 20–22 g) obtained from the Experimental Animal Center of China Medical University (Shenyang, China) were housed in 50–60% humidity and kept on a 12-h light/dark cycle at 22±2°C with free access to food and water. The experimental protocol was approved by the Animal Ethics Committee of The Fourth Affiliated Hospital of China Medical University.

DE was induced as described previously (26). Briefly, the mice were randomly divided into four groups: Control (con); DE model (DE); DE with balanced salt solution treatment (DE+vehicle); and DE with 2% rebamipide treatment (DE+Reb) (n=10 mice/group). The mice in the three DE groups were exposed to an environment of <20% relative humidity, 21–23°C and airflow of 15 l/min for 16 h/day, and received subcutaneous injections of scopolamine hydrobromide (0.5 mg/0.2 ml resolved in saline; Dalian Meilun Biotech Co., Ltd., Dalian, China) at 8:00 a.m., 11:00 a.m., 2:00 p.m. and 5:00 p.m. daily for 2 weeks. The mice in the control group were placed in a normal environment without airflow and received subcutaneous injections of an equal volume of saline at the same time points. Following the establishment of the model, the mice in the DE+Reb group were treated with 2% rebamipide eye drops (Mucosta^®; Otsuka Pharmaceutical Co., Ltd.) four times daily for 2 weeks. The mice in the DE+vehicle group were treated with salt solution at the same time points.

Tear secretion was evaluated using the phenol red thread test. The mice were fixed in a 50-ml tube with the bottom cut off and phenol red-impregnated cotton threads (Tianjin Jingming New Technological Development Co., Ltd., Tianjin, China) were applied to the lateral canthus for 15 sec. The length of wetting of the thread, which turned red, was measured (mm).

The stability of the tear film was examined using a tear film breakup time (BUT) assay and corneal fluorescein staining. The mice were anesthetized with 50 mg/kg pentobarbital sodium (intraperitoneal) and 1 µl 1% fluorescein sodium (Alcon Laboratories, Inc., Duluth, GA, USA) were applied to the conjunctival sac. The BUT was recorded as the duration between the natural blink response of the mice to the appearance of a dark area, representing tear film breakup, using slit-lamp biomicroscopy and a cobalt blue light (BX900; Haag-Streit AG, Koeniz, Switzerland). The BUT was recorded three times and the average time was used. Then excess fluorescein sodium was rinsed using PBS and the corneal epithelium deficit was evaluated using fluorescein punctate staining the duration of which depended on the natural blink response time of each mouse. The staining was scored according to previously established standards (27).

From each group, six mice were randomly selected for histological examination. Following rebamipide administration, the mice were sacrificed by deep anesthesia and the cornea and lacrimal gland were collected and fixed in 10% neutral formalin at 4°C for 24 h. The fixed samples were embedded in optimal cutting temperature (OCT) medium and cut into 8-µm-thick cryosections. The sections were stained with hematoxylin for 5 min and eosin for 3 min at room temperature. The histological alterations in the cornea and lacrimal gland were observed under a light microscope at ×400 or ×200 magnification (DP73; Olympus Corporation, Tokyo, Japan).

ELISA was used to measure the levels of IL-1β, IL-6, tumor necrosis factor-α (TNF-α), IL-10, IL-17, and IL-23 in the conjunctiva and lacrimal gland. The tissues were homogenized in PBS at 4°C and subjected to three freeze-thaw cycles. The homogenates were centrifugated at 10,000 × g at 4°C for 20 min. The supernatant was collected for measurement. The measurements were performed using commercial ELISA kits [MultiSciences (Lianke) Biotech Co., Ltd., Hangzhou, China], according to the manufacturer's protocols. The catalog number of each kit was: IL-1β: EK201B2/2; IL-6: EK2062/2; TNF-α: EK2822/2; IL-10: EK2102/2; IL-17: EK2172/2; IL-23: EK2232.

Total RNA was isolated from the conjunctiva and lacrimal gland samples using the RNApure total RNA Fast Isolation kit (BioTeke Corporation, Beijing, China). The isolated RNA was quantified and reverse transcribed in a 20 µl reaction system containing 1 µg RNA sample, 1 µl random primer, 2 µl dNTP, 1 µl oligo(dT)₁₅, 4 µl 5XBuffer, 0.5 µl RNasin and 1 µl Moloney murine leukemia virus reverse transcriptase. Reverse transcription reactions were carried out at 42°C for 50 min and 95°C for 5 min. RT-qPCR was performed using 2X Power Taq PCR Master Mix (BioTeke) and SYBR Green (Beijing Solarbio Science & Technology Co., Ltd., Beijing, China) on an Exicycler 96 real-time PCR system (Bioneer Corporation, Daejeon, Korea). All reactions were incubated at 94°C for 10 min, 60°C for 20 sec, and 72°C for 30 sec, followed by 40 cycles of 72°C for 2 min 30 sec, 40°C for 5 min 30 sec, 60°C for 30 sec, and 25°C for 1 min. The primers used are listed in Table I. The relative expression of the genes was measured against β-actin and was calculated using the 2^−ΔΔCq method (28).

The fixed samples were embedded in OCT medium and cut into 10-µm-thick cryosections. The sections were boiled in a citrate buffer in a microwave oven for 10 min for antigen retrieval. Following blocking in 10% goat serum (Beijing Solarbio Science & Technology Co., Ltd.) for 15 min at room temperature, the sections were incubated with anti-keratin 10 (K10; 1:50; cat. no. sc-53252), or anti-IL-17 (1:50; cat. no. sc-374218) (both Santa Cruz Biotechnology, Inc., Dallas, TX, USA) and anti-cluster of differentiation (CD)4 (1:100; cat. no. 19068-1-AP; Proteintech Group, Inc., Chicago, IL, USA) antibodies or with anti-forkhead box P3 (Foxp3; 1:50; cat. no. sc-53876; Santa Cruz Biotechnology, Inc.) and anti-CD4 (1:100) antibodies at 4°C overnight. Following washing, the sections were incubated with fluorescein isothiocyanate-conjugated goat/anti-mouse (1:200; A0568; Beyotime Institute of Biotechnology, Haimen, China) and Cy3-conjugated goat/anti-rabbit secondary antibodies (1:200; A0516; Beyotime Institute of Biotechnology) for 60 min at room temperature. The staining was observed under a BX50 fluorescence microscope (Olympus Corporation) at ×400 magnification.

The data are expressed as the mean ± standard deviation. Experiments were repeated 6 or 10 times as described above. Groups were compared using one-way analysis of variance followed by Tukey's post hoc test for data with a normal distribution, and by Games-Howell analysis with Dunn's multiple comparisons test for skewed data using SPSS 19.0 software (IBM Corp., Armonk, NY, USA). P<0.05 was considered to indicate a statistically significant difference.

The effects of rebamipide on the symptoms of DE were evaluated by the examination of corneal fluorescein staining, BUT and tear production. A total of 4 weeks after the model was established, the mean corneal fluorescein staining score was significantly increased in the DE group (8.10±1.37) compared with the control group (1.30±0.48) (P<0.01). Vehicle treatment for 2 weeks did not affect the score (7.90±1.37; P>0.05 vs. the control group). However, following 2 weeks of treatment with 2% rebamipide, the corneal fluorescein staining score decreased to 3.5±0.71, which was significantly lower compared with that in the DE+vehicle group (P<0.01) (Fig. 1A and B).

Regarding the measurement of tear film BUT, the results revealed that the values taken from the DE group were significantly lower compared with those from the control group (2.80±1.22 vs. 12.70±1.42; P<0.01), and rebamipide treatment partly restored this change (5.70±1.16; P<0.05 vs. the DE+vehicle group) (Fig. 1C). In accordance with the results from the BUT measurement, the tear volume (represented as the wetting length of the phenol red thread) in the DE group was markedly reduced compared with the control group (0.51±0.17 vs. 4.50±0.29; P<0.01), and rebamipide treatment slightly but significantly increased the tear production in DE mice (0.93±0.17; P<0.01 vs. the DE+vehicle group). The tear production in the vehicle group demonstrated no statistically significant difference when compared with that in the DE group (0.54±0.16; P>0.05) (Fig. 1D).

Hematoxylin and eosin staining revealed that superficial epithelial cells in the corneas of the control mice were well arranged and tightly attached, and the surface of the cornea was smooth (Fig. 2A). However, the epithelial layers of the cornea were disrupted, and the epithelial cells were desquamated in the DE mice. In addition, the lacrimal glands in the control group exhibited a unbroken structure and uniform size of the acinus (Fig. 2B). In the DE mice, the lacrimal glands exhibited notable vacuolization and a disordered cellular arrangement. The vehicle did not affect the histological alterations in the lacrimal glands, whereas 2% rebamipide administration markedly attenuated these pathological alterations.

In the conjunctival sections, the number of goblet cells was decreased in the DE mice compared with that in the control mice (Fig. 3). Rebamipide treatment significantly increased the number of goblet cells. Furthermore, epidermis-specific K10 expression was examined in the corneal epithelia. In the control group, few K10-positive epithelial cells were observed in the cornea, whereas DE induced a marked increase in the expression of K10, which is a hallmark of squamous metaplasia. Rebamipide treatment significantly weakened the intensity of K10 expression compared with that observed in the vehicle-treated group (Fig. 3).

To further investigate the alterations in inflammatory-associated factors following rebamipide treatment, the levels of known pro- and anti-inflammatory cytokines in the lacrimal gland and corneal tissues were determined. As demonstrated in Fig. 4, the mRNA expression levels and protein content of the pro-inflammatory cytokines IL-1β, IL-6, TNF-α, IL-17 and IL-23 were significantly enhanced, whereas the anti-inflammatory cytokine IL-10 was significantly decreased, in the samples from DE mice. Treatment with 2% rebamipide for 2 weeks markedly reversed these effects.

To evaluate the impact of rebamipide on the ratio of Th17 and Tregs in the ocular microenvironment, double-immunofluorescence was performed to locate CD4⁺/IL-17⁺ and CD4⁺/Foxp3⁺ double-positive cells in the cornea and lacrimal gland. As demonstrated in Fig. 5, more CD4⁺/IL-17⁺ and CD4⁺/Foxp3⁺ cells were observed in the cornea and lacrimal gland specimens from the DE mice compared with those of the control, indicating that DE altered the ratio of Th17/Tregs in the ocular microenvironment. Treatment with rebamipide (2%) led to decreases in the numbers of CD4⁺/IL-17⁺ cells and CD4⁺/Foxp3⁺ cells.