Adult New Zealand white rabbits (1.5–2.0 kg) were purchased from the Qinglong Mountain Experimental Animal Center (No. SCXK Jiangsu 2024-0001; Nanjing, China) and housed in the Pharmacology Animal Laboratory of Jiangsu Province Hospital of Traditional Chinese Medicine (Jiangsu, China). Housing conditions were maintained at a constant temperature of 22 °C $\pm$ 2 °C, relative humidity of 60% $\pm$ 5%, and a 12-hour light/dark cycle, with ad libitum access to food and water. Experimental rabbits were humanely euthanized via intravenous injection of sodium pentobarbital (120 mg/kg; 50 mg/mL) into the marginal ear vein. Death was confirmed by $\ge$3 minutes of absent heartbeat (auscultation), loss of corneal reflex, and fixed dilated pupils. All procedures were performed by trained personnel in a dedicated room to minimize animal stress.

The DED model was established by subcutaneous injection of Scop at the nape of the neck (Fig. 1A) at a dose of 2.0 mg/kg per injection, administered four times daily (8:00, 11:00, 14:00, and 18:00) for 21 consecutive days. Successful model establishment was confirmed by strong positive corneal fluorescein (FL) staining (Fig. 1B).

Fig. 1.

Schematic diagram of the experimental principle. (A) Adult New Zealand white rabbits received four daily subcutaneous injections of Scop for 21 consecutive days. From day 22, EA stimulation was applied to bilateral ocular acupoints Jingming (BL-1), Cuanzhu (BL-2), Sizhukong (TE-23), Tongziliao (GB-1), Taiyang (EX-HN5). And $\alpha$-BGT was injected via the ear vein for 14 days. (B) Experimental groups included the Vehicle control group, Scop-treated group, Scop-treated + sham EA group, and Scop-treated + EA group. By day 21, all groups except the Vehicle control group showed strong positive corneal FL staining, confirming successful establishment of the DED model (The magnification of the slit lamp lens is 10$\times$). Scale bar = 1 mm. (C) SIT was measured in New Zealand white rabbits. EA, electroacupuncture; $\alpha$-BGT, $\alpha$-Bungarotoxin; FL, fluorescein.

Experiment 1: Male adult New Zealand white rabbits were randomly divided into four groups (n = 6 per group): (1) Vehicle control group; (2) Scop group; (3) Scop + sham group; and (4) Scop + EA group. Experiment 2: Adult New Zealand white rabbits of both sexes were randomly divided into five groups (n = 6 per group): (1) Vehicle control group; (2) Scop group; (3) Scop + EA group; (4) Scop + $\alpha$-BGT + EA group; and (5) Scop + $\alpha$-BGT group.

EA treatment was initiated on day 22 and administered once daily for 14 consecutive days. For the Scop + sham EA group, blunt needle puncture at acupoints was performed daily from day 22 for 14 days. A selective $\alpha$7nAChR antagonist, $\alpha$-Bungarotoxin ($\alpha$-BGT, pk-ca707-00010-1; Germany), was administered via the ear vein at a dose of 4.0 µg/kg daily from day 22 for 14 days. The vehicle control group received subcutaneous injections of 0.9% sodium chloride according to the same schedule and dosage as the Scop-treated groups for a total of 35 days.

EA stimulation was performed using the Hwato Electronic Acupuncture Stimulator SDZ-IIB Health Care Device (Nanjing Jisheng Medical Technology Co., Ltd., Nanjing, China). Rabbits were restrained in a specialized rabbit fixation box. Bilateral ocular acupoints, including Jingming (BL-1), Cuanzhu (BL-2), Sizhukong (TE-23), Tongziliao (GB-1), and Taiyang (EX-HN5), were selected for EA stimulation. Acupuncture needles (0.3 mm diameter, 15 mm length; Hwato) were inserted subcutaneously to a 5 mm depth. EA was delivered with a sparse-dense wave (2 Hz/20 Hz), 0.2 ms pulse width, and 1 mA intensity (adjusted to elicit slight muscle twitching at the insertion site). The positive electrode of the EA device was connected to Jingming (BL-1), and the negative electrode to the ipsilateral Taiyang (EX-HN5). EA stimulation was applied for 20 min per day.

A corneal staining filter paper strip was placed into the lower eyelid fornix of the rabbit and moistened. FL was evenly distributed over the cornea through blinking. Corneal epithelial damage was graded using a cobalt blue filter. The cornea was divided into four quadrants, with scoring criteria as follows: 0 (no staining), 1 (5 staining spots), 2 ($>$5 staining spots), and 3 (large-area FL plaques). The total score was the sum of scores from the four quadrants, with a maximum of 12 points.

After moistening the FL sodium staining paper, it was gently touched to the conjunctival fornix of the lower eyelid 2–3 times. The rabbit was encouraged to blink several times to facilitate complete diffusion and distribution of the dye across the ocular surface. Under a slit lamp (Chongqing Shang Bang Medical Equipment Co., Ltd., Chongqing, China), the timer recorded the time of the first corneal dry spot, repeatedly measured three times, and the average time was used.

One end of the Schirmer strip was folded and inserted into the space between the middle and outer one-third of the lower eyelid and left in place for 5 min. The wet length of the strip was recorded after the strip was taken out (Fig. 1C).

Tear samples (20–50 µL) were collected from the lateral palpebral fissure (lacrimal lake) using a glass microcapillary pipette. Tear osmolarity was measured using the Vapro 5600 Osmometer (Wescor Inc., South Logan, UT, USA).

After the experimental protocol, animals were euthanized for the collection of corneas, lacrimal glands, and conjunctivae. These excised tissues were subsequently fixed in 4% paraformaldehyde, embedded in paraffin, sectioned at 5 µm, dried, and stained with hematoxylin and eosin. Histological changes were assessed under light microscopy.

A portion of the conjunctival tissue was excised, rinsed thoroughly with physiological saline, and weighed. Then the tissue was homogenized under ice bath conditions. Subsequently, the homogenate was diluted with 300 µL physiological saline and centrifuged, after which the supernatant was collected. A cholinesterase inhibitor (neostigmine, 10 µM) was added during tissue collection/homogenization. Samples were kept on ice, and assays were completed within 2 h of homogenization. The concentrations of ACh (JEB-14586; Nanjing Jin Yibai, China) and $\alpha$7nAChR (JEB-14612) were determined using specific assay kits. The levels of ACh and $\alpha$7nAChR were quantified using an enzyme-linked immunosorbent assay reader.

Conjunctival tissues were processed for embedding and sectioning, followed by antigen retrieval and blocking. Sections were incubated with a primary anti-HMGB1 antibody (1:500, GB11103-100; Servicebio, Wuhan, China) at 4 °C overnight, followed by incubation with horseradish peroxidase (HRP)-conjugated goat anti-rabbit IgG secondary antibody (1:1000, RS0002; ImmunoWay, San Jose, CA, USA) and staining with 3,3^′-diaminobenzidine. Nuclei were counterstained with hematoxylin. Using Image-Pro Plus 6.0 software (National Institutes of Health, Bethesda, MD, USA), the integrated optical density (IOD) of positive staining and pixel area were measured, and the areal density was calculated as IOD/area.

Conjunctival paraffin sections were dewaxed through a standard xylene-ethanol gradient and rehydrated. After antigen retrieval (citrate buffer, pH 6.0, 95 °C for 20 min) and blocking in 5% goat serum (room temperature, 30 min), sections were divided into two groups for single-labeling detection. Then the sections were incubated overnight at 4 °C with primary antibodies against HMGB1 (1:300, GB11103-100; Servicebio) and $\alpha$7nAChR (1:300, 144730; LSBio, Seattle, WA, USA). After three washes with phosphate-buffered saline (5 min each), the sections were incubated with Alexa Fluor® 488-conjugated goat anti-rabbit IgG H&L (1:1000, ab150077; Abcam, Cambridge, MA, USA) at room temperature for 1 h. Nuclei were counterstained with DAPI for 5 min. Sections were mounted with anti-fade mounting medium and observed under a fluorescence microscope (Olympus, Tokyo, Japan), with images captured using a consistent exposure time.

Conjunctival paraffin sections were dewaxed and stained with Periodic-acid Schiff reagent to highlight intracellular stored mucus and the secretory products of goblet cells. Positively stained goblet cells in the conjunctiva were counted, and the distance between the first and last goblet cells was measured to calculate the average number of goblet cells per millimeter.

Initial fixation was conducted in 0.1 mol/L phosphate buffer (pH 7.4) with 2.5% glutaraldehyde, followed by post-fixation in 1% osmium tetroxide. After dehydration through an ascending series of alcohols, the samples were embedded in epoxy resin. Then, small blocks (1 mm³) were excised from the central region of the conjunctiva. These blocks were subjected to double staining with lead acetate and uranyl acetate, after which they were observed under a transmission electron microscope.

Equal amounts of cell lysates were extracted from the conjunctiva, and the protein concentration was determined using the BCA protein assay. Protein was resolved by 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis, electrotransferred to a PVDF membrane (Millipore, Burlington, MA, USA), blocked in 5% non-fat dry milk, and incubated overnight at 4 °C with the following primary antibodies: anti-HMGB1 rabbit polyclonal antibody (1:1000, GB11103, Servicebio) , anti-$\alpha$7nAChR rabbit monoclonal antibody (1:1000, 144730, LSBio), and GAPDH rabbit monoclonal antibody (1:1000, 5174S; Cell Signaling Technology, Danvers, MA, USA) for immunoblotting. After three washes with Tris-buffered saline containing 0.05% Tween 20 for 10 min each, the membrane was incubated with HRP-conjugated goat anti-rabbit IgG (1:1000, RS0002; ImmunoWay) for 1 h at room temperature. Band intensities were analyzed using Quantity One software v4.6.6 (Bio-Rad, Hercules, CA, USA).

Total protein was extracted from the conjunctiva, and the protein concentration was measured using the BCA protein assay. A rabbit cytokine quantification array (QAL-CYT-1 kit; RayBiotech, Inc., Norcross, GA, USA) was employed to detect the expression of 10 cytokines (interleukin 1$\alpha$ [IL-1$\alpha$], IL-1$\beta$, IL-8, IL-17A, IL-21, leptin, macrophage inflammatory protein-1beta, matrix metalloproteinase 9 [MMP-9], neural cell adhesion molecule 1 [NCAM-1], tumor necrosis factor alpha [TNF-$\alpha$]) in the conjunctiva, and the assay was repeated three times. Then the signal was scanned using a Cy3 excitation curve with the InnoScan 300 Microarray Scanner (Innopsys, Carbonne, France). Data analysis was performed using QAL-CYT-1 data analysis software (Q-Analyzer).

All experimental procedures were performed in at least three independent replicates. All data are presented as the mean $\pm$ standard error of the mean. All continuous data were first tested for normality using the Shapiro-Wilk test and for homogeneity of variance using Levene’s test. For data that conformed to a normal distribution and exhibited homogeneous variance, one-way analysis of variance (ANOVA) was used to compare the overall differences among multiple groups. If the main effect among groups was significant, further post-hoc analysis was performed with Bonferroni correction. For within-group comparisons, paired t-tests with Bonferroni correction were applied. Statistical significance was established for findings where both the uncorrected p-value was less than 0.05, and the corrected p-value remained less than 0.05 following adjustment for multiple comparisons. SPSS 22.0 software (IBM SPSS Inc., Armonk, NY, USA) was used for statistical analyses. p 0.05 was considered statistically significant.

To determine whether EA stimulation can alleviate ocular surface functional abnormalities resulting from DED, a series of assessments was conducted following 3 and 5 weeks of continuous subcutaneous injection of Scop. At 3 and 5 weeks, the FL was elevated in the Scop and Scop + Sham EA groups compared to the Vehicle control group, indicating severe corneal epithelial damage in these Scop-treated groups. Concurrently, the BUT was significantly reduced in all three groups relative to the vehicle control group (Fig. 1B and Fig. 2B,D). After 5 weeks of Scop injection, the tear osmolarity was markedly increased, while the SIT was significantly decreased in the Scop and Scop + Sham EA groups (Fig. 2A,B,D,F,G). EA stimulation for 2 weeks counteracted these adverse effects. These results suggest that EA exerts protective effects against corneal epithelial damage, consistent with the improved epithelial layer integrity observed in the histological sections (Fig. 2C). Additionally, EA alleviated ocular surface damage and mitigated ocular functional abnormalities induced by DED, with specific benefits in reducing tear osmolarity and enhancing tear film stability.

Fig. 2.

EA stimulation alleviates ocular surface tissue damage and functional impairment in DED. (A) Representative images of corneal FL staining on day 35 (The magnification of the slit lamp lens is 10$\times$). Scale bar = 1 mm. (B) Quantitative analysis of corneal FL staining scores (n = 10, Both eyes). (C) Representative hematoxylin and eosin (H&E) staining images of the cornea on day 35 (n = 3). Scale bars = 625 µm or 200 µm. (D) Quantitative analysis of the tear BUT (n = 10, Both eyes). (E) Representative H&E staining images of the lacrimal gland on day 35 (n = 3). Scale bars = 625 µm or 200 µm. (F) Quantitative analysis of tear flow (SIT, n = 10, Both eyes). (G) Quantitative analysis of tear osmolarity (n = 6). Data are presented as the mean $\pm$ standard error of the mean (SEM). Between-group comparisons: one-way ANOVA with Bonferroni correction. Within-group comparisons: paired t-tests with Bonferroni correction. *p 0.05, **p 0.01, ***p 0.001, ****p 0.0001. DED, dry eye disease; BUT, break-up time; SIT, Schirmer I test; ANOVA, one-way analysis of variance.

The impact of EA stimulation on the mitigation of ocular surface tissue damage caused by DED was assessed by histopathological analysis. After 5 weeks of Scop injection, the corneal epithelial layers were diminished with superficial squamous epithelial cells sloughing off, accompanied by edema in the stromal layer and visible corneal epithelial shedding in the Scop and Scop + Sham EA groups. EA stimulation notably improved the corneal damage, evidenced by an increase in corneal layers, a less disordered structure of the superficial squamous epithelium, and a reduction in corneal epithelial shedding and stromal edema (Fig. 2A,C). The lacrimal glands, which primarily secrete the aqueous component of tears, displayed acinar atrophy and increased cavity size compared to the Vehicle control group in the Scop and Scop + Sham EA groups. EA stimulation attenuated the acinar atrophy (Fig. 2E). The conjunctival epithelial cells showed disrupted layers and polarity, with infiltration of lymphocytes and neutrophils, and partial epithelial cell detachment in the Scop and Scop + Sham EA groups. EA stimulation similarly ameliorated the conjunctival epithelial cells, with a decrease in epithelial cell detachment and a reduction in lymphocytes (Fig. 3A). Conjunctival goblet cells secrete conjunctival mucin to form the mucin layer of the tear film, which is pivotal for maintaining tear film stability and ocular surface homeostasis [23, 24]. In comparison to the Vehicle control group, the Scop and Scop + Sham EA groups notably decreased the number of conjunctival goblet cells, whereas EA stimulation increased the number of goblet cells (Fig. 3B,C).

Fig. 3.

EA stimulation modulates conjunctival tissue and ultrastructure in DED. (A) Conjunctival H&E staining on day 35 (n = 3). Scale bar = 625 µm. (B,C) Periodic-acid Schiff (PAS) staining of conjunctival goblet cells and counting of goblet cells in each group (n = 3). Scale bar = 200 µm. (D) Transmission electron microscopy of conjunctival epithelial cells, depicting microvilli (Mv), mitochondria (M), RER, nuclei (N), nucleoli (Nu), desmosomes (De), goblet cells (GCs), and secretory granules (SGs). Scale bars = 1 µm or 500 nm. Quantitative data are presented as the mean $\pm$ SEM, Between-group comparisons: one-way ANOVA with Bonferroni correction. *p 0.05, **p 0.01, ****p 0.0001. RER, rough endoplasmic reticulum.

The ultrastructural integrity of conjunctival epithelial cells was assessed in the Vehicle control group, revealing a pattern of short-columnar arrangement with uniform cytoplasmic density, continuous cell membrane structures, and the presence of microvilli along the membranes. The mitochondrial membranes displayed elevated density, and the rough endoplasmic reticulum (RER) exhibited no signs of dilation, with ribosomes firmly anchored to its surface. Intercellular bridges were observed, and no significant widening of the intercellular spaces was noted. A profusion of secretory granules was evident adjacent to the goblet cells. By contrast, the Scop group displayed flattened cells with necrotic outermost layers. The cell membranes had microvilli, but the cytoplasm was sparse and disintegrating with a diminished number of organelles. Mitochondria were swollen, the membrane appeared blurred, and the matrix and cristae were slightly sparse. Intermediate layer cells were in a state of contraction, with a higher cytoplasmic density and increased intercellular space. The Scop + Sham EA group exhibited flattened cells, with microvilli attached to the membranes, and nuclei were elongated, with minimal accumulation of heterochromatin at the margins and a blurred nuclear membrane. Mitochondria were swollen, with a slightly sparse matrix and cristae, and the RER showed slight dilation with sparse ribosomes on its surface. Deeper-layer cells appeared contracted, with widened intercellular spaces. The Scop + EA group showed cells arranged in a short-columnar pattern, with uniform distribution of the cytoplasm. The cell membrane was intact, but microvilli were sparsely distributed. Nuclei showed a slight concavity, with heterochromatin aggregating in small patches and a clear nuclear membrane. Mitochondria were swollen, with intact membranes and slight dissolution of the matrix and cristae. The Golgi apparatus was not noticeably hypertrophied, and intercellular bridges were visible, with some local intercellular spaces slightly widened (Fig. 3D).

Previous experimental studies have established the pivotal role of $\alpha$7nAChR in the cholinergic anti-inflammatory pathway [25]. To determine the involvement of $\alpha$7nAChR in the anti-inflammatory effects of EA stimulation, we quantified the expression levels of ACh in each group following EA stimulation. Following EA stimulation, compared to the Scop group, the expression of $\alpha$7nAChR and ACh in conjunctival tissue was significantly upregulated, whereas Sham acupuncture did not elicit an increase (Fig. 4C,D). HMGB1 is released during inflammatory responses, and its excessive release can lead to further cell damage and the release of HMGB1, exacerbating the inflammatory cascade. Immunofluorescence and Western blotting were utilized to detect $\alpha$7nAChR and HMGB1 in conjunctival epithelial tissue, which are pivotal regulators of inflammatory factors. Scop-induced conjunctival epithelial tissue displayed reduced $\alpha$7nAChR expression and increased HMGB1 expression, which EA stimulation effectively reversed (Fig. 4A,B,E).

Fig. 4.

EA stimulation suppresses Scop-induced conjunctival inflammation and activates $\alpha$7nAChR. (A,B) The levels of $\alpha$7nAChR and HMGB1 in conjunctival epithelial tissue were evaluated using immunofluorescence staining, Scale bar = 60 µm. (E) Western blotting (n = 3), respectively. (C,D) Expression of $\alpha$7nAChR and ACh in conjunctival tissue was quantified using ELISA (n = 6). (F,G) A rabbit conjunctival protein array was utilized to analyze changes in cytokines and chemokines in the conjunctiva (n = 6), Cytokine circular heatmap (F), Expression of cytokines and chemokines in the conjunctiva (n = 6) (G). Quantitative data are presented as the mean $\pm$ SEM. Between-group comparisons: one-way ANOVA with Bonferroni correction. *p 0.05, **p 0.01, ***p 0.001, ****p 0.0001. $\alpha$7nAChR, alpha-7 nicotinic acetylcholine receptor; HMGB1, high mobility group box 1; ELISA, Enzyme-Linked Immunosorbent Assay.

The conjunctiva represents the most metabolically active region of the ocular surface, serving as a critical interface for the immune response to a wide array of exogenous insults, including allergens, pathogens, and toxins. This mucosal epithelium forms the first line of defense in the ocular surface’s innate and adaptive immune systems [26]. To elucidate the mechanism by which EA stimulation modulates inflammatory responses, we employed an antibody array to quantify the expression of a panel of 10 cytokines and chemokines. Compared to the Vehicle control group, the Scop-treated conjunctiva displayed pronounced upregulation of IL-1$\alpha$, IL-1$\beta$, IL-8, IL-17A, IL-21, leptin, MMP-9, NCAM-1, and TNF-$\alpha$. However, following EA stimulation, elevation in the expression of IL-1$\alpha$, IL-1$\beta$, IL-8, MMP-9, NCAM-1, and TNF-$\alpha$ was significantly attenuated, signifying the marked inhibition of inflammation by EA stimulation, a response that was absent in the Scop + Sham EA group (Fig. 4F,G).

EA stimulation exerts its anti-inflammatory effects by stimulating the $\alpha$7nAChR, which in turn suppresses the generation of pro-inflammatory cytokines within the conjunctival tissue. To investigate the role of $\alpha$7nAChR in the EA-mediated anti-inflammatory response, we employed the specific $\alpha$7nAChR antagonist $\alpha$-BGT to block its activity. Intravenous injection of $\alpha$-BGT via the ear vein reversed the EA-induced upregulation of $\alpha$7nAChR expression, while the expression of ACh remained unaltered (Fig. 5A,B). Additionally, we re-evaluated the ocular surface phenotype of the Scop-induced dry eye rabbits following $\alpha$-BGT treatment. This treatment abrogated EA stimulation’s ability to (1) reduce corneal FL staining (Fig. 5C,E), (2) prolong tear BUT (Fig. 5F), (3) increase tear flow (Fig. 5H), (4) reduce tear osmolarity (Fig. 5I), and (5) mitigate corneal and lacrimal gland tissue damage (Fig. 5D,G). These data collectively indicate that EA stimulation elicits a localized activation of $\alpha$7nAChR, which subsequently suppresses ocular surface inflammation and functional impairment.

Fig. 5.

EA stimulation mediates its anti-inflammatory effects through $\alpha$7nAChR. (A,B) Expression of $\alpha$7nAChR and ACh in conjunctival tissue was quantified using the enzyme-linked immunosorbent assay (n = 6). (C) Representative images of corneal FL staining on day 35, black arrow: Intense fluorescence indicates corneal epithelial defects (The magnification of the slit lamp lens is 10$\times$). Scale bar = 1 mm. (D) Corneal H&E staining on day 35, black arrow: Corneal epithelial detachment and epithelial structural disorder; Blue arrow: Corneal stromal edema (n = 3). Scale bars = 625 µm or 100 µm. (E) Corneal FL staining score (n = 6). (F) Tear BUT (n = 6). (G) Lacrimal gland H&E staining on day 35 (n = 3). Scale bars = 625 µm or 200 µm. (H) Tear flow measurement (n = 6). (I) Tear osmolarity (n = 5). Quantitative data are presented as the mean $\pm$ SEM. Between-group comparisons: one-way ANOVA with Bonferroni correction. Within-group comparisons: paired t-tests with Bonferroni correction. *p 0.05, **p 0.01, ***p 0.001, ****p 0.0001.

To elucidate the potential cholinergic anti-inflammatory mechanism underlying EA stimulation, we examined whether the expression of HMGB1 and related pro-inflammatory cytokines and chemokines was inhibited following $\alpha$-BGT blockade of $\alpha$7nAChR expression. We observed that $\alpha$-BGT reversed the beneficial effects of EA stimulation on conjunctival tissue damage, with a thinner epithelial layer and a corresponding decrease in the number of conjunctival goblet cells (Fig. 6A,B,F). Immunofluorescence and immunohistochemical assays were employed to assess the expression of $\alpha$7nAChR and HMGB1 in conjunctival epithelium. $\alpha$-BGT antagonist inhibited the EA-induced increase in $\alpha$7nAChR expression, leading to an increase in HMGB1 expression (Fig. 6C–E,G). Although EA treatment resulted in a significant reduction in the levels of IL-1$\alpha$, IL-1$\beta$, IL-8, MMP-9, NCAM-1, and TNF-$\alpha$ in the Scop g roup, these anti-inflammatory effects were not observed in the $\alpha$-BGT group. Conversely, following treatment with the $\alpha$-BGT antagonist, the inhibitory effects of EA on the total expression levels of IL-1$\beta$, IL-8, MMP-9, and TNF-$\alpha$ were not completely abolished (Fig. 6H,I). These findings indicate that EA suppresses the HMGB1 pathway by enhancing $\alpha$7nAChR expression in conjunctival tissue to control inflammation. However, the regulatory effects of EA on IL-1$\beta$, IL-8, MMP-9, and TNF-$\alpha$ are not entirely mediated by the $\alpha$7nAChR-dependent signaling pathway.

Fig. 6.

EA stimulation engages the $\alpha$7nAChR-mediated HMGB1 pathway in conjunctival tissue, suppressing inflammation. (A) H&E staining of the conjunctiva on day 35 (n = 3). (B) PAS staining for conjunctival goblet cells. (C,D) Immunofluorescence assays to determine the expression levels of $\alpha$7nAChR and HMGB1 in the conjunctival epithelial layer. (E) Immunohistochemical analysis for assessing HMGB1 expression in the conjunctival epithelial tissue. (F) Quantification of goblet cell counts across different groups (n = 3). (G) Western blotting to evaluate the protein expression of $\alpha$7nAChR and HMGB1 in the conjunctival epithelial tissue (n = 3). (H) Heatmap depicting the alterations in cytokine and chemokine profiles in rabbit conjunctival cells, as detected by protein microarray (n = 6). (I) Expression profiling of conjunctival cytokines and chemokines (n = 6). Statistical data are expressed as the mean $\pm$ SEM. Between-group comparisons: one-way ANOVA with Bonferroni correction. Scale bar = 200 µm. Statistical significance is indicated by *p 0.05, **p 0.01, ***p 0.001, and ****p 0.0001.