This article is mentioned in:

Introduction

Diabetes mellitus (DM) has become a global public health crisis, affecting 537,000,000 individuals worldwide in 2021 (1) and has been identified as a risk factor for dry eye (DE) disease (2,3). Clinical studies have found that over half of patients with diabetes have DE disease, which progresses as the duration of diabetes increases (4,5). Diabetic DE disease is characterized by decreased corneal sensitivity, reduced tear secretion, altered tear proteins and corneal epithelial lesions (6–9). Hyperglycemia impairs corneal epithelial basal cells and the basement membrane (10), but the specific mechanism by which corneal epithelial lesions form remains unclear. Therefore, understanding the molecular mechanism of corneal epithelial dysfunction in diabetic DE disease is desirable.

Mucin1 (MUC1) is a member of the mucin family, which contains O-glycosylated proteins that play key roles in the tear film homeostasis and protect the ocular surface (11,12). MUC1 is expressed across the cornea and conjunctival epitheliums in humans but only in the corneal epithelium in mice (13,14). The extracellular domain of the N-terminal alpha subunit of MUC1 (MUC1-ND) is responsible for cell adhesion and the cytoplasmic tail of MUC1 (MUC1-CT) is involved in cell signaling (15). Previous studies have reported that the MUC1-CT carries out extensive functions beyond its ability to interact with several proteins, such as Src, catenin β1 (CTNNB1) and ErbBs (16,17). For example, MUC1-CT promotes cell adhesion by binding CTNNB1/β-catenin and junction plakoglobin/γ-catenin. Additionally, MUC1-CT facilitates RAB27A-mediated exosome secretion by interacting with the pocket of RAB27A (18). Studies have also found that MUC1 is extracellularly cleaved within the sperm protein, enterokinase and agrin (SEA) module in MUC1-ND by ADAM17 or MT1MMP (19,20). Miyazaki et al (21) show that MUC1-ND functions as a hydrophilic barrier that hinders the entry of lipophilic anticancer drugs, leading to chemoresistance. Increasing evidence suggests that MUC1 is also implicated in inflammatory diseases (22,23). However, the function of MUC1 and the underlying molecular mechanism involved in diabetic DE disease remain unclear.

Transient receptor potential cation channel subfamily V member 1 (TRPV1), a nociceptive sensor, is associated with ocular surface pain (24). TRPV1 enhances the excitability of nociceptive sensors activated by heat, protons and inflammatory mediators (25). TRPV1 channel dysfunction is shown to induce decreased corneal sensitivity, which is correlated with the pathogenesis of DE disease (6). Notably, TRPV1 is found to be translocated to astrocyte membranes to promote migration and inflammatory infiltration in models of neonatal hypoxic-ischemia and hypoxic hypoglycemia (26,27), but it is involved in the protection of diabetic rats after hypoxic preconditioning (28), suggesting that TRPV1 plays a complex role in the inflammatory process. The TRPV1 channel activation is also shown to induce the proliferation and migration of human corneal epithelial cells (29). MUC1 interacts with TRPV5 through galectin 3, enhancing renal TRPV5 activity and preventing kidney stone formation (30). In addition, MUC1 decreases the internalization of Ca-selective TRP channels TRPV5 and TRPV6 while promoting their presence on the cell surface, aiding in maintaining Ca2+ balance (31). These findings highlight the close relationship between MUC1 and TRP channels. However, whether MUC1 interacts with TRPV1 and the underlying signaling mechanism in corneal epithelial cell proliferation remain to be elucidated.

The present study found decreased proliferation of mouse corneal epithelial cells (MCECs), accompanied by decreased levels of MUC1-ND and TRPV1, in diabetic DE mice. In addition, capsazepine (CPZ; a TRPV1 antagonist) restricted the proliferation of MCECs. There was an interaction between MUC1-ND and the TRPV1 protein in the control group but not in the diabetic DE group. Furthermore, the activation of the AKT signaling pathway was attenuated in diabetic DE mice and it was downstream of TRPV1. In summary, the results illustrated that the MUC1-ND/TRPV1/AKT axis plays an essential role in the proliferation of MCECs and provided new insight into the pathogenesis of corneal epithelial dysfunction in diabetic DE disease.

Materials and methods

Reagents

Primary antibodies against the following were provided by Abcam: MUC1-ND (cat. no. ab45167), AKT1/2/3 (cat. no. ab185633), p-AKT (Thr308; cat. no. ab38449), Bcl2 (cat. no. ab182858), β-Actin (cat. no. ab8227) and β-tubulin (cat. no. ab6046). Primary antibodies against the following were purchased from Proteintech Group, Inc.: MUC1-CT (cat. no. 23614-1-AP), TRPV1 (cat. no. 66983-1-Ig) and phosphorylated (p-)AKT (Ser473; cat. no. 28731-1-AP). Primary antibodies against the following were purchased from Cell Signaling Technology, Inc.: Caspase3 (cat. no. 9662) and IgG (cat. no. 2729). The primary antibodies of CK3+12 (cat. no. Bs-2369R) and TRPV1 (cat. no. Bs-23926R) were purchased from BIOSS. The following secondary antibodies were obtained from Invitrogen (Thermo Fisher Scientific, Inc.): Alexa Fluro 488 (goat anti-mouse, cat. no. a11029; goat anti-rabbit, cat. no. a11008) and Alexa Fluro 594 (goat anti-mouse, cat. no. a11032; goat anti-rabbit, cat. no. a11012). The following secondary antibodies were obtained from Abcam: goat anti-rabbit IgG/HRP (cat. no. ab6721) and goat anti-mouse IgG/HRP (cat. no. ab6789). CPZ (cat. no. HY15640) and MK2206 (cat. no. HY-10358) were purchased from MedChemExpress.

Animals

Male C57BL/6 mice at aged of 6–8-weeks and weighing 18–22 g were purchased from the Experimental Animal Center of Southern Medical University (Guangzhou, China). Mice were cared for in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research (32). All experimental procedures received approval from the Institutional Animal Care and Use Committee of Southern Medical University (Guangzhou, China; approval no. LAEC-2021-032). The mice were housed in a specific pathogen-free rated animal vivarium (temperature, 20–26°C; light/dark cycle, 12/12 h; humidity: 40–70%) and the research team monitored the animals twice daily. Health assessments included monitoring weight (twice weekly), food and water intake and a general evaluation of activity levels, panting and fur condition. A total of 155 mice were used, of which 149 were euthanized using carbon dioxide (CO2) inhalation. The absence of a heartbeat and dilation of pupils for a period of ≥5 min were verified in order to confirm mortality; if these conditions were not met, cervical dislocation was then performed. Sacrificed mice were used for corneal and conjunctival histopathology, as well as for the primary culture of mouse corneal epithelial cells. Of the diabetic group, six succumbed unexpectedly, probably due to complications associated with diabetes. The duration of the experiment ranged from 1.5–2.5 months. The ocular surface parameter measurements were conducted in a non-invasive and painless manner. Additionally, 3% sodium pentobarbital was administered intraperitoneally at a dose of 50 mg/kg to induce sedation in the mice.

Diabetic DE mouse model

A streptozotocin (STZ; MilliporeSigma)-induced diabetic mouse model was established and Na-citrate buffer administered to the control group. To induce DM, mice were fasted for 4 h and injected with a low dose of STZ (50 mg/kg) for 5 consecutive days. Mice were fed a normal diet (control group) or a high-fat/high-sugar diet (DM group). Corneal fluorescein staining (CFS), tear ferning (TF) tests and phenol red thread tests were performed before treatment and 7, 30 and 60 days after the final injection. Corneal sections were conducted hematoxylin and eosin (H&E) staining on days 30 and 60 after the final injection and conjunctival sections were stained with periodic acid Schiff (PAS) on days 30 and 60 after the final injection. After the aforementioned evaluation, a diabetic DE mouse model was established and the mice were divided into a control group and a diabetic DE group.

CFS

To estimate the defects of corneal epithelium in the mice, 1 µl of 1% liquid sodium fluorescein was dropped into the lateral conjunctival sac. After 3 min, the eyes were imaged with a slit-lamp microscope (SL-17; Kowa Optimed, Inc.) under cobalt blue light. A total of four quadrants of cornea were scored by two independent blinded individuals using a standardized four-scale system: 0 points for no staining; 1 point for <30 stained dots; 2 points for >30 non-diffuse stained dots; 3 points for severe diffuse staining but no positive plaques; and 4 points for positive fluorescein plaques. The total score was the sum of four quadrants scores (0–16).

TF tests

Tear samples (0.5 µl) collected from each eye of the mice were dried on a microscopic glass slide at room temperature and <40% humidity for 20 min. The TF patterns were imaged under a Leica bright field microscope (Leica Microsystems GmbH; magnification, ×10) and analyzed using a new 5-point grading scale (33) from 0 to 4; a grade higher than grade 2 indicated possible DE disease.

Tear production

The amount of tears was quantified by the phenol red thread (Zone-Quick; Showa Yakuhin Kako Co., Ltd.). A smooth tweezer placed the folded end of the thread into the temporal one-third of the lateral conjunctival sac for 15 sec. After 15 sec, the wetting length of thread (red portion) was measured in millimeters.

Histopathology

The eyeball and eyelid tissues of the mice were excised and fixed in 4% paraformaldehyde (PFA) at 4°C overnight. The tissues were dehydrated using varying concentrations of ethanol, followed by transparency treatment with xylene, immersion in wax, and embedding in paraffin. Paraffin sections were cut into 4 µm sagittal sections with a microtome (RM2245; Leica Microsystems GmbH). Frozen sections were sliced into sagittal sections (6–10 µm thick) with a microtome (CM1950; Leica Microsystems GmbH). H&E or PAS were performed on the paraffin sections, while the frozen sections underwent immunofluorescence (IF) staining.

To evaluate defects in the corneal epithelium, eye sections were stained with hematoxylin for 5 min and eosin for 1 min at room temperature. To evaluate the goblet cell density of the conjunctiva, PAS staining of the conjunctiva sections was performed using a PAS kit (Beijing Solarbio Science & Technology Co., Ltd.). The eye sections were stained with periodic acid for 5 min and Schiff for 10 min at room temperature. Images of the staining sections were captured under a microscope (DM2500; Leica Microsystems GmbH). The thickness of the corneal epithelium in captured images was measured with ImageJ 1.8 (National Institutes of Health). The goblet cells in the superior and inferior conjunctiva were counted by two independent blinded individuals in three sections of each eye. The average number of the three sections was recorded as the goblet cell density of the eye.

Isolation and culture of primary mouse corneal epithelial cells (MCECs)

MCECs were isolated by enzymatic digestion. The mice from the control group and diabetic DE group were sacrificed by CO2. Animals were placed into CO2 anesthesia box with ≥7.5% CO2 and after the animals gradually lost consciousness, the CO2 concentration was raised to 100% until respiratory cardiac arrest occurred. Their eyes were surgically extracted, placed into a 10 cm dish (Corning Life Sciences) and washed with PBS three times. Under a Leica S9 stereomicroscope (Leica Microsystems GmbH), the lens and posterior eyecups were discarded to detach the corneas, which were dissected into four equal pieces. Afterward, the corneas were transferred into a new 10 cm dish containing 0.25% EDTA-trypsin (Gibco; Thermo Fisher Scientific, Inc.) and digested at 37°C for 10 min. The control group cells were maintained in DMEM/F12 (Gibco; Thermo Fisher Scientific, Inc.) containing 10% FBS, 1% nonessential amino acids (Gibco; Thermo Fisher Scientific, Inc.), insulin (5 µg/ml; MilliporeSigma) and 100 U/ml penicillin/streptomycin (Gibco; Thermo Fisher Scientific, Inc.). The diabetic DE group cells were cultured in DMEM with high glucose and the same supplements used to maintain the control group.

CCK8 assay

Cell Counting Kit-8 (CCK8; Gibco; Thermo Fisher Scientific, Inc.) solution was used to detect the cell proliferative capacity of each group or treated cells. Primary MCECs from each group were cultured at 2,000 cells per well in 96-well plates (Corning Life Sciences) for 3 days. CCK8 solution (10 µl) was added to each well at 4, 24, 48 and 72 h and then incubated at 37°C under 5% CO2 for 2 h. The optical density (OD) was measured at 450 nm wavelength via a microplate reader (BioTek; Agilent Technologies, Inc.) and the cell growth curves were plotted.

To evaluate the effect of CPZ or MK2206 on cell viability, primary MCECs from each group were treated with DMSO-soluble CPZ (0, 10, 30 µM) or water-soluble MK2206 (1 µM) for 24 h. After the addition of CCK8 solution and incubation for 1.5 h, the OD was measured and the cell viability was calculated.

Cell IF staining

Primary MCECs were stained with cytokeratin 3+12 (CK3+12) for cell identification. The primary MCECs were cultured on 6-well plates (Corning Life Sciences) with round cover slips (12 mm) for 24 h. Then, the cells were first fixed in 4% PFA for 15 min at room temperature and blocked with 5% BSA (Thermo Fisher Scientific, Inc.) in PBS. Following incubation with the primary antibodies anti-CK3+12 (1:200), anti-MUC1-CT (1:200) and anti-TRPV1 (1:200) and fluorescent secondary antibodies (1:500), the cells were counterstained with DAPI for 5 min at room temperature (cat. no. ab285390; Abcam). Images were captured with an inverted confocal microscope (Nikon Corporation).

IF colocalization analysis

Frozen corneal sections were fixed in cold acetone for 10 min at 4°C. After blocking with 5% BSA in PBS for 1 h at room temperature, they were incubated with anti-MUC1-ND (1:200) and anti-TRPV1 (1:300) overnight at 4°C and stained with secondary antibodies (1:500) for 1 h at room temperature, then counterstained with DAPI for 5 min at room temperature. Images were captured with a Nikon Ti-E microscope (Nikon Corporation) equipped with an A1R confocal system. IF colocalization analysis was conducted with ImageJ 1.8 (National Institutes of Health).

Western blotting

Primary MCECs from each group or treated cells were lysed in RIPA (Thermo Fisher Scientific, Inc.) lysis buffer mixed with 1% phosphate inhibitor cocktail 100X (Gibco; Thermo Fisher Scientific, Inc.) and 1% protease inhibitor cocktail 100X (Gibco; Thermo Fisher Scientific, Inc.). A BCA protein assay kit (Nanjing KeyGen Biotech Co., Ltd.) was used to examine the protein concentration. Equal amounts of protein samples were loaded to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and moved to polyvinylidene fluoride membranes (MilliporeSigma). A total of 30 µg protein/lane was loaded, and 5% stacking gel and 8, 10 and 12% separating gel was used. The membranes were blocked with 5% BSA for 2 h at room temperature and incubated with primary antibodies against the following overnight at 4°C: p-AKT (Ser473) (1:1,000), MUC1-CT (1:800), MUC1-ND (1:1,000), TRPV1 (1:1,000), AKT1/2/3 (1:1,500), p-AKT (Thr308) (1:800), Bcl2 (1:2,000), caspase3 (1:1,000) and β-Actin (1:1,000). After incubation with secondary antibodies (1:2,000) for 1 h, the bands were detected by an enhanced chemiluminescence (ECL) system (Bio-Rad ChemiDoc XRS+; Bio-Rad Laboratories, Inc.). Quantification of the band intensity was conducted with ImageJ 1.8 (National Institutes of Health).

Co-immunoprecipitation (Co-IP) assays

Co-IP assays were performed by using a Pierce cross-linked bead immunoprecipitation/coprecipitation kit (cat. no. 88805; Thermo Fisher Scientific, Inc.). First, primary MCECs were lysed with IP lysis buffer and centrifuged at 10,000 × g for 10 min at 4°C. Then, the lysates with 500 µg proteins were immunoprecipitated with 5 µg anti-TRPV1 (1:20), 5 µg anti-MUC1-ND (1:20) or 1 µg IgG (1:50) coupled to protein A/G magnetic beads at room temperature for 1.5 h. The beads were washed in IP buffer and eluted in IP elution buffer for 10 min. All incubations were taken on a rotator or mixer. The beads were washed to remove unbound proteins, and a low pH elution buffer was used to separate bound antigen from the antibody cross-linked beads. Neutralization buffer was included to prevent precipitation of the isolated antigen and to ensure protein activity in downstream applications. Western blotting was conducted to detect the bound proteins.

Statistical analysis

All data are shown as the mean ± standard deviation. Statistical tests were performed by SPSS 20.0 (IBM Corp.) software and graphs were prepared by Prism 8.0 (GraphPad; Dotmatics) software. Student's t-test was used to compare differences of two groups. Comparisons between more than two groups were made by one-way ANOVA followed by the LSD test or Welch test with Dunn's multiple-comparisons test. P<0.05 was considered to indicate a statistically significant difference.

Results

Establishment of a diabetic DE mouse model

To establish a diabetic DE mouse model, STZ-induced diabetic mice were created on a high-fat and high-sugar diet and phenol red thread tests, CFS, TF tests and histological examinations were conducted to evaluate DE disease.

First, STZ was used to induce diabetes in mice and the mice were then followed for 60 days following the final STZ injection to determine changes in the ocular surface as the duration of diabetes increased (Fig. 1A). Mice in the DM group were confirmed to have higher blood glucose levels, lower body weights and increased food intake compared with those of mice in the control group (Table I), indicating that the diabetic mice had been successfully created. In addition, the mice were observed at baseline and 7, 30 and 60 days following treatment by phenol red thread, CFS and TF tests (Fig. 1B). As illustrated in Fig. 1C, the tear production was similar level between the control and DM groups at baseline and on days 7 after treatment. However, the diabetic mice exhibited decreased tear production on days 30 and 60 after STZ injection. These results demonstrated that the lacrimal gland function in secreting tear fluids was impaired in the DM group on days 30 and 60 after STZ injection.

Figure 1. STZ-induced diabetic mice and evaluation of DE disease. (A and B) Experimental protocol for diabetic mice preparation and assays to evaluate DE disease at various time points. (C) Quantitative analysis of tear production using the phenol red thread test. (D) CFS was imaged by a slit-lamp microscope under cobalt blue light on days 7, 30 and 60 after final injection. Green-stained areas represent corneal epithelial defects. (E) Quantitative analysis of corneal epithelial damage. (F) Representative images captured during the TF test at baseline and days 7, 30 and 60 after final injection. Scale bar, 100 µm. (G) Quantitative analysis of TF pattern grades. In C, E and G, the mean values ± standard deviation are shown (n=20 eyes, 10 animals per group). A two-sided Student's t test was used to calculate the statistical significance. ns, not significant, *P<0.05, **P<0.01, ***P<0.001. STZ, streptozotocin; DE, dry eye; CFS, corneal fluorescein staining; TF, tear ferning.

Table I. Blood glucose, weight and food intake of mice before and after treatment.

Table I.

Blood glucose, weight and food intake of mice before and after treatment.

Group		Control	Diabetes mellitus
Blood glucose (mmol/l)	Baseline	6.24±0.78	5.18±0.57
	7 days	6.80±0.84	24.83±1.94a
	30 days	7.72±0.62	25.49±1.83a
Body weight (g)	Baseline	20.18±0.64	20.35±0.71
	7 days	21.28±0.64	21.81±0.78
	30 days	26.96±0.81	23.61±1.08a
Food intake (g/day)	Baseline	4.70±0.43	5.05±0.50
	7 days	5.55±0.71	8.75±0.58a
	30 days	6.16±0.58	8.83±0.46a

{ label (or @symbol) needed for fn[@id='tfn1-mmr-30-6-13337'] } The data are expressed as means and standard deviation (n=10).

a P<0.001.

CFS (Fig. 1D) of the DM group after STZ injection showed scattered punctate staining on days 7, severe diffuse staining on days 30 and positive fluorescein plaque staining on days 60. However, no staining of the corneas of the control group was observed. The DM group mice had a higher CFS score than the control group mice (Fig. 1E). After treatment, corneal epithelial defects in the diabetic mice increased on days 30 and days 60 following STZ injection.

The diabetic mice exhibited decreased TF and poor TF pattern grades on days 7, 30 and 60 after treatment (Fig. 1F and G). At baseline, there was no significant difference between the two groups. On 7, 30 and 60 days after treatment, the TF grade of the DM group was significantly higher than that of the control group. These findings suggested that in the diabetic mice, the composition and proportions of tear proteins, especially mucins, were already changed.

Furthermore, histopathologic examination of the cornea and conjunctiva were performed. During the 60-day follow-up, it was found that compared with the control mice, the diabetic mice exhibited a thinner corneal epithelium with reduced cell layers and the disturbance of basal cells (Fig. 2A). The thickness of the corneal epithelium in H&E-stained sections (Fig. 2B) was significantly smaller in the DM group compared with the control group on days 30 and 60 (P<0.01, P<0.01), suggesting that DM had injured the corneal epithelium. As shown in Fig. 2C and D, the diabetic mice had a lower goblet cell density than the control mice on days 30 and days 60 after treatment, illustrating that DM had impaired the number and function of conjunctival goblet cells.

Figure 2. Histopathology of the ocular surface tissues. (A) Representative images of corneal sections following hematoxylin and eosin staining. Scale bar, 50 µm. (B) Statistical analysis of the thickness of the corneal epithelium, as measured in images of (A). (C) Representative images of conjunctival sections stained with PAS. Scale bar, 200 µm. (D) The goblet cell density in PAS-stained sections. In (B) and (D), data are show as mean ± standard deviation (n=5 animals per group). A two-tailed Student's t-test was used to calculate the statistical significance. ns, not significant, *P<0.05, **P<0.01, ***P<0.001. PAS, periodic acid Schiff; DM, diabetes mellitus.

After the aforementioned evaluation, a diabetic DE mouse model was established. Taken together, the results indicated that the diabetic mice exhibited decreased tear secretion, increased corneal epithelial damage and poorer TF grades, accompanied by corneal epithelium and conjunctiva injury, proving that the diabetic DE mouse model had been created.

Isolation, culture and identification of primary MCECs

To explore the mechanism of corneal epithelial lesions in the diabetic DE mice in vitro, primary corneal epithelial cells were cultured. The corneas of mice in the control and diabetic DE groups were isolated and primary MCECs from the two groups by cultured enzymatic digestion (Fig. 3A). The morphology of primary MCECs in the two groups cultured on day 10 is shown in Fig. 3B. The MCECs grew by static adherence and the cells in the two groups had similar shapes and resembled spindles, ovals and polygons. The cells in the control group grew rapidly, but those in the diabetic DE group grew slowly and were sparse. This result indicated the poor proliferation of cells in the diabetic DE group.

Figure 3. Isolation, culture and identification of primary MCECs in the control and diabetic DE groups. (A) Flow chart of the isolation and culture of primary MCECs by enzymatic digestion in the control and diabetic DE groups. (B) Morphology of primary MCECs in the two groups cultured on days 10 by enzymatic digestion. Scale bar, 100 µm. (C) Representative IF images demonstrating the distribution of CK3+12 in primary MCECs from the two groups. Scale bar: 100 µm. (D) Western blot analysis showed CK3+12 protein expression in primary MCECs from the two groups. (E) Densitometry analysis of the data in (D). In (E), data are shown as mean ± standard deviation from three independent experiments. ns, not significant as calculated by two-tailed Student's t-test. MCECs, mouse corneal epithelial cells; DE, dry eye; IF, immunofluorescence.

The cells were subjected to IF staining for CK3+12, a marker of corneal epithelial cells. The CK3+12 protein distributions in the two groups were similar, as illustrated in Fig. 3C. The CK3+12 protein levels in the cells from the two groups were did not significantly differ (Fig. 3D and E). These results identified the cultured cells as corneal epithelial cells.

Hyperglycemia impairs cell proliferation, accompanied by decreased MUC1-ND and TRPV1

To further elucidate the functional changes in primary MCECs, CCK8 assays were conducted, which demonstrated attenuated cell proliferation in the diabetic DE group (Fig. 4A). The expression of MUC1-CT and TRPV1 in primary MCECs of the control and diabetic DE groups was detected and the protein localization was assessed by IF staining (Fig. 4B). The difference in TRPV1 expression between cells in the control and diabetic DE groups was visible, but a difference in MUC1-CT expression was not. The colocalization of MUC1-CT and TRPV1 was also observed in the membrane and cytoplasm. In addition, it was observed that TRPV1 and MUC1-ND were significantly downregulated at the protein level in cells of the diabetic DE group by western blotting and densitometry analysis (Fig. 4C and D). However, the MUC1-CT protein levels of the two groups of cells did not significantly differ.

Figure 4. Hyperglycemia impairs cell proliferation, accompanied by decreased MUC1-ND and TRPV1 levels. (A) CCK8 assays showed the cell proliferative ability of the two groups. The OD value at 450 nm was recorded. (B) IF staining was used to locate and detect the expression of MUC1-CT and TRPV1 in primary MCECs from the control and diabetic DE groups. Scale bar, 100 µm. (C) Western blot analysis of TRPV1, MUC1-ND and MUC1-CT in primary MCECs. β-Actin was used as a loading control. (D) Densitometry analysis of the data in (C). In (A) and (D), data are shown as mean ± standard deviation from three or more independent experiments. ns, not significant, *P<0.05 vs. the corresponding control, as analyzed by two-tailed Student's t test. MUC1-ND, MUC1 extracellular domain; TRPV1, transient receptor potential cation channel subfamily V member 1; IF, immunofluorescence; MCECs, mouse corneal epithelial cells; DE, dry eye; MUC1-CT, cytoplasmic tail of MUC1; MUC1, mucin1.

These results indicated that hyperglycemia impaired cell proliferation in the diabetic DE mice, accompanied by reduced MUC1-ND and TRPV1 levels and that TRPV1 and MUC1 may interact in primary MCECs.

MUC1-ND binds TRPV1 to positively regulate the proliferation of MCECs MUC1-ND interacts with TRPV1

To further validate the interaction between MUC1-ND and TRPV1, IF colocalization analysis of frozen corneal sections from the two groups and Co-IP analysis of the two groups of cells were conducted. The distribution and expression of TRPV1 and MUC1-ND are shown in Fig. 5A. IF staining confirmed that TRPV1 and MUC1-ND colocalized in the cell membrane and cytoplasm of the corneal epithelium. The present study also observed reduced TRPV1 and MUC1-ND expression in corneal tissues from the diabetic DE mice in comparison with the control mice. IF colocalization analysis of IF staining images were conducted by ImageJ software to obtain Pearson's correlation coefficients, which were used for statistical analysis. As illustrated in Fig. 5B and C, the Pearson r between MUC1-ND and TRPV1 in the diabetic DE group was 0.60±0.12, which was significantly lower than that in the control group (0.82±0.13). This result indicated that the colocalization of MUC1-ND and TRPV1 was reduced in the diabetic DE mice. Additionally, Co-IP analysis of the two groups of cells was carried out and the association of TRPV1 and MUC1-ND was observed (Fig. 5D). The present study found evidence that MUC1-ND interacted with TRPV1 in the control cells. By contrast, TRPV1 did not coimmunoprecipitate with MUC1-ND in the diabetic DE cells. The present study also tried IP:TRPV1 to immunoblot mucin1-CT and mucin1-ND to immunoblot TRPV1 (Fig. S1), it found that TRPV1 did not interact with mucin1-CT in the control or diabetic DE cells, which may be related to the binding site between the two proteins. These results illustrated that the interaction between MUC1-ND and TRPV1 in the corneal epithelium was decreased in the diabetic DE mice, which may be related to the downregulation of MUC1-ND and TRPV1 in the diabetic DE mice.

Figure 5. TRPV1 interacts with MUC1-ND to promote cell viability. (A) Frozen corneal sections were stained for TRPV1 and MUC1-ND in the control and diabetic DE groups. Three channels (red, blue and green, right panel) were used for IF analysis. n=5 animals per group. Scale bar, 50 µm. (B) Scatter diagram from IF colocalization analysis of the data in (A) by ImageJ software. (C) Statistical analysis of the Pearson r between TRPV1 and MUC1-ND in the corneal epithelium in the two groups. n=5 animals per group. (D) Coimmunoprecipitation and western blotting showed that TRPV1 bound MUC1-ND in primary MCECs from the two groups. (E) CCK8 assays detected the cell viability of the two groups after treatment with CPZ (0, 10 and 30 µM), a TRPV1 antagonist. In (C) and (E), data are shown as mean ± standard deviation from at least three independent experiments. ns: not significant, *P<0.05, **P<0.01, ***P<0.001 vs. the corresponding control, as analyzed by either two-tailed Student's t test for two groups or by one-way ANOVA with LSD's multiple-comparison test for more than two groups. TRPV1, transient receptor potential cation channel subfamily V member 1; MUC1-ND, MUC1 extracellular domain; MUC1, mucin1; DE, dry eye; IF, immunofluorescence; MCECs, mouse corneal epithelial cells; CPZ, capsazepine.

TRPV1 positively regulates cell proliferation

To verify the regulatory role of TRPV1 in cell proliferation, primary MCECs of the control and diabetic DE groups were treated with CPZ (0, 10 and 30 µM), a TRPV1 antagonist. Cell viability was detected by CCK8 assays, as shown in Fig. 5E. The viability in control group cells treated with CPZ at 30 µM was significantly lower than that of cells treated without CPZ treatment (P<0.01) and cells treated with CPZ at 10 µM (P<0.05). However, there was no significant difference in cell viability between the untreated cells and cells treated with 10 µM CPZ. In the diabetic DE group, the lowest cell viability was found in cells treated with CPZ at 30 µM, followed by those treated with CPZ at 10 µM and those without CPZ treatment. The viability of cells treated with 10 µM CPZ and 30 µM CPZ was significantly lower than that of cells without CPZ treatment (P<0.001, P<0.001). However, the cell viability of cells treated with 10 µM CPZ was similar to that of cell treated with 30 µM CPZ. Collectively, these results demonstrated that CPZ, a TRPV1 antagonist, restricted cell proliferation as its concentration increased.

These data suggested that MUC1-ND interacts with TRPV1, which can positively regulate cell proliferation.

TRPV1 promotes cell proliferation by partly activating the AKT signaling pathway

To clarify the mechanism by which the binding of MUC1-ND and TRTV1 regulates cell proliferation, western blotting and CCK8 assays were conducted. Primary MCECs in the control and diabetic DE groups were treated with CPZ (0, 10 and 30 µM) for 24 h and then harvested for western blotting. CPZ was formulated with DMSO as a solvent, so CPZ 0 µM meant cells treated with DMSO in the control or diabetic DE groups. As shown in Fig. 6A, there was decreased expression of TRPV1 in control cells treated with CPZ at 30 µM and diabetic DE cells treated with CPZ at 0, 10 and 30 µM, indicating that TRPV1 expression was inhibited by CPZ. As shown in Fig. 6A and B, similar levels of AKT1/2/3 in the control groups were observed but in diabetic DE groups, it was significantly increased in 30 µM CPZ group compared with 0 µM CPZ group. The level in the diabetic DE + 10 µM CPZ group was similar to the other two diabetic DE groups. The expression of p-AKT(Ser473) decreased significantly with CPZ treatment in a dose-dependent manner in the control groups and the lowest level was observed in the diabetic DE + 30 µM CPZ group (Fig. 6A and C). The expression in diabetic DE + 10 µM CPZ group was not significantly different from the other two diabetic DE groups. The expression of Bcl2 decreased with CPZ treatment in a dose-dependent manner in the control and diabetic DE groups. The protein level of caspase3 increased in diabetic DE groups compared with the control groups with different CPZ treatment. It was more abundant in the control + 30 µM CPZ group than in control groups treated with 0 and 10 µM CPZ. Similar levels was found in the three diabetic DE groups. These results suggested that CPZ, a TRPV1 antagonist, attenuated the AKT signaling pathway.

Figure 6. TRPV1 positively regulates cell proliferation by partly activating the AKT pathway. (A) Western blotting showing the protein expression of TRPV1, AKT1/2/3, p-AKT (Ser473), p-AKT (Thr308), caspase3 and Bcl2 in the two groups of cells upon treatment with the TRPV1 antagonist CPZ (0, 10 and 30 µM) for 24 h. (B and C) Densitometry analysis of AKT1/2/3, p-AKT (Ser473) in (A). (D) Western blotting and (E) densitometry analysis of the protein expression of AKT1/2/3, p-AKT (Ser473), p-AKT (Thr308), caspase3 and Bcl2 in primary MCECs of the control and diabetic DE groups. (F) CCK8 assays showed the cell viability in the control group without CPZ treatment and upon treatment with CPZ at 30 µM or with CPZ at 30 µM + MK2206 (an AKT inhibitor) at 1 µM for 24 h. Western blotting showing the protein expression of TRPV1, p-AKT (Ser473), caspase3 and Bcl2 in primary MCECs of the diabetic DE group treated as described in (G). In (B, C and F), quantified data are shown as mean ± standard deviation from three or more independent experiments. ns, not significant, *P<0.05, **P<0.01, ***P<0.001 vs. the corresponding control, by either two-tailed Student's t-test for two groups or by one-way ANOVA followed by LSD multiple-comparison test for more than two groups. TRPV1, transient receptor potential cation channel subfamily V member 1; p-, phosphorylated; CPZ, capsazepine; DE, dry eye; STZ, streptozotocin.

In addition, western blotting was performed to detect the protein expression of AKT1/2/3, p-AKT (Ser473), p-AKT (Thr308), caspase3 and Bcl2. As illustrated in Fig. 6D and E, the protein levels of p-AKT (Ser473) and Bcl2 were significantly reduced in the diabetic DE group, but the protein levels of AKT1/2/3, p-AKT (Thr308) and caspase3 was significantly increased in the diabetic DE group when compared with the control group. The results indicated that activation of the AKT signaling pathway was attenuated in the diabetic DE mice.

Furthermore, cells in the control group were treated with DMSO, 30 µM CPZ or treated with 30 µM CPZ + 1 µM MK2206 (an AKT inhibitor) for 24 h. Then, CCK8 assays were conducted to detect cell viability and western blotting. As shown in Fig. 6F, the cell viability of the 30 µM CPZ-treated group was significantly lower than that of the untreated group but was not significantly different from that of the CPZ + MK2206-treated group. The cell viability of the CPZ + MK2206-treated group was significantly lower than that of the DMSO group (P<0.001). These results demonstrated that AKT signaling pathway inhibition was partly responsible for the effect of CPZ (a TRPV1 antagonist) on cell proliferation. As shown in Fig. 6G, the protein expressions of TRPV1, p-AKT (Ser473) and Bcl2 were reduced in other two groups compared with that of the DMSO group. The caspase3 protein level was increased in the 30 µM CPZ-treated group and the CPZ + MK2206-treated group.

The aforementioned results suggested that TRPV1 positively regulated cell proliferation by partly activating the AKT signaling pathway.

Discussion

The present study investigated the biological role of the MUC1-ND/TRPV1/AKT axis in the moderation of corneal epithelial cells in diabetic mice with DE disease. First, diabetic DE model mice were established with significantly increased corneal epithelial defects, reduced tear production, poor TF pattern grades and an impaired ocular surface structure. Second, the expression levels of MUC1-ND and TRPV1 were found to be significantly reduced in the corneal epithelium of diabetic DE mice. Third, TRPV1 was shown to positively regulate the proliferation of MCECs. Notably, MUC1-ND was shown to interact with the TRPV1 protein in the control group but not in the diabetic DE group. Furthermore, it was found that TRPV1 positively regulated cell proliferation by partly activating the AKT signaling pathway. In general, the present study revealed that the interaction between MUC1-ND and TRPV1 partly activated the AKT signaling pathway to promote MCEC proliferation.

DE disease is a multifactorial disorder and its most distinguished feature is the loss of homeostasis of the tear film (2). The establishment of DE animal models is necessary to study the fundamental mechanisms of DE disease, which is nearly identical in animals and in humans (34). Previous studies have confirmed that DE animal models, including chemical- or surgery-induced DE models, environmental factor-induced DE models, genetically engineered DE models and models prepared with combined methods, are reliable and effective in imitating ocular surface microenvironments (34,35). Common parameters used to evaluate DE disease in mouse models are CFS to assess surface structure, the TF test to assess tear proteins, the phenol red thread test to assess tear secretion and histopathological examination (34,35). In the present study, the diabetic mice were followed for 60 days and it was confirmed that they exhibited significantly increased corneal epithelial defects, decreased tear production, poor TF pattern grades, impaired corneal epithelial structure and a lower goblet cell density on day 30 following the final STZ injection. The method used to establish the diabetic DE model was similar to that used in another previous study (36). The diabetic DE model mice were identified to be reliable tools in the study of pathophysiological mechanisms and the underlying molecular cascade in diabetic DE disease. Notably, it was also observed that TF was lessened and even disappeared in diabetic mice with DE, indicating that the tear proteins differed in diabetic mice with DE disease. Then, primary MCECs from mice in the control and diabetic DE groups were cultured and identified, consistent with a previous study (37).

MUC1 is a key member of the membrane-associated mucin (MAM) family with an important barrier function (15). MUC1 accumulates with other MAMs on the corneal epithelia and it is cross-linked by galectin-3 to form an interlocking lattice for pathogen exclusion (14). Damage to the ocular surface epithelium and barrier disruption causes ocular surface diseases, including DE disease. One clinical study reported that MUC1 expression levels were decreased in the conjunctival epithelium of patients with DE disease (38). MUC1 is also confirmed to be cleaved at its extracellular domain (MUC1-ND) (19,20). Part of the soluble extracellular subunit of MUC1 can be detected in human tears (39). In the present study, MUC1-CT expression was found decreased in diabetic DE mice, but the difference was not significant compared with that of the control mice, whereas MUC1-ND levels were significantly decreased in the corneal epithelial cells of diabetic DE mice compared with control mice. This suggested that MUC1 was cleaved and that its soluble extracellular large subunit (MUC1-ND) was shed from the corneal epithelial cell surface in diabetic mice with DE disease. The present study revealed that MUC1 abnormalities may be closely linked with corneal epithelial homeostasis in diabetic DE disease.

Growing evidence shows the downregulation of TRPV1 in colorectal, endometrial, renal and skin cancers but the upregulation of TRPV1 in the U373 glioblastoma line, high-grade astrocytes and the RT4 renal cell carcinoma line, indicating that TRPV1 mediates the balance between cell proliferation and apoptosis (40). TRPV1 activation acts as a double-edged sword in that it induces both improvements in cell proliferation and migration in corneal injury (41–43) and the release of proinflammatory cytokines in human corneal epithelial cells stimulated by hyperosmolarity (25,44,45). Previous studies (30,31,46) have also shown that MUC1 regulates inflammation via interactions with several different signaling pathways and correlates with the stabilization of ion channels, which are essential to corneal epithelial homeostasis (15). MUC1 forms a lattice with TRPV5 via galectin-3 to increase renal TRPV5 activity to protect against nephrolithiasis (30). The present study first confirmed the downregulation of TRPV1 in the corneal epithelia of diabetic DE mice compared with control mice. It next analyzed the relationship between TRPV1 levels and corneal cell proliferation. It was found that CPZ, a TRPV1 antagonist, restricted proliferation in primary MCECs, indicating that TRPV1 played a key role in promoting MCEC proliferation in diabetic DE disease. The present study further investigated the interaction between MUC1 and TRPV1 and the results demonstrated that MUC1-ND interacted with the TRPV1 protein in the control group but not in the diabetic DE group. These results demonstrated that the interaction between MUC1-ND and TRPV1 positively promoted corneal epithelial cell proliferation. This interaction was attenuated in the diabetic mice and may contribute to corneal epithelial dysfunction in DE disease.

The Ser and Thr kinase AKT, also known as protein kinase B (PKB), has become the focus in biological and medicinal studies, together with its function and regulation in diabetes (47). AKT activation occurs downstream of inositol phosphate 3-kinase (PI3K), a lipid kinase associated with cell transformation and the insulin response. The activation of PI3K leads to the phosphorylation of two key residues on AKT1, catalyzing the activation of Thr308 in the T loop and Ser473 in the C-terminal hydrophobic motif. AKT activation is greatly decreased when Ser473 is unphosphorylated, which stabilizes Thr308 phosphorylation and AKT activation. Attenuated activation of the AKT signaling pathway inhibits corneal epithelial wound healing in a model of corneal epithelial trauma in diabetic mice (48). Some studies have found that TRPV1 channel activation can result in the phosphorylation of AKT for signal transduction (29,49). The present study found that the AKT signaling pathway was attenuated in diabetic mice with DE disease, accompanied by the decreased phosphorylation of AKT (Ser473). Additionally, it was found that CPZ, a TRPV1 antagonist, reduced the activation of the AKT signaling pathway. However, this inhibition was partly responsible for the effect of CPZ (a TRPV1 antagonist) on cell proliferation. Collectively, these findings revealed that TRPV1 positively regulated cell proliferation by partly activating the AKT signaling pathway. Notably, the present study revealed the interaction between MUC1-ND and TRPV1 and its regulation in the corneal epithelial cells of diabetic DE mice. In the present study it was found that MUC1-ND acts as a protein partner that interacts with TRPV1, which positively regulates MCEC proliferation by partly activating the AKT signaling pathway. The decrease in this interaction in the diabetic mice may have induced corneal epithelial dysfunction in the progression of diabetic DE disease.

There was limitation to the present study. MUC1-ND in the tears of diabetic mice failed to be detected via western blotting, due to very little tear secretion from the mice. It may be an improvement to collect the tear fluid of diabetic patients with dry eye disease to detect the mucin1-ND level. Relevant clinical trials are required to assess the role of MUC1-ND and TRPV1 in diabetic DE disease.

In summary, the present study illustrated that the interaction of MUC1-ND with TRPV1 promoted MCEC proliferation by partly activating the AKT signaling pathway. The findings provided new insight into the pathogenesis of corneal epithelial dysfunction in diabetic DE disease.

Supplementary Material

Supporting Data

Acknowledgements

Not applicable.

Funding

The present study was supported by Natural Science Foundation of Guangdong Province, China (project no. 2022A1515012346) and the Science and Technology Program of Guangzhou, China (project no. 202002020046).

Availability of data and materials

The data generated in the present study may be requested from the corresponding author.

Authors' contributions

HQL, XHL, YZ and YYZ contributed to the study design. HQL, YZ, YTC, RZ, XHL, YYZ, WKZ, HC, JH and SFF conducted experiments. HQL and YZ performed statistical analysis and drafted the manuscript. HQL and YZ confirm the authenticity of all the raw data. All authors read and approved the final manuscript.

Ethics approval and consent to participate

All experimental procedures were approved by the Institutional Animal Care and Use Committee of Southern Medical University (Guangzhou, China; approval no. LAEC-2021-032) and conducted in compliance with the ARVO statement for the Use of Animals in Ophthalmic and Vision Research (32).

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

References