The present study was approved by the Ethics Committee of the Affiliated Hospital of Shandong University of Traditional Chinese Medicine (AWE-2022-055). All the animal studies were performed in strict accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Three-color short-haired healthy guinea pigs (2 weeks old, male, and 110–130 g) were supplied by Danyang Changyi Experimental Animal Breeding Co., Ltd. In an animal room with a constant temperature of 20–25 °C, all animals were reared in clean, transparent plastic cages (5–6 per cage) under 300 lux light for a 12/12 h day and night cycle, with feeding conditions. Before enrollment, all experimental animals received ocular examination by a computerized refractometer (TOPCON, KR8900, Japan) and a hand-held retinal camera (Japan Kowa Co., Ltd., Genesis-D) to eliminate spontaneous myopia, cataracts, or corneal disease. After a 1-week adaptive feeding period, the study cohort of 120 healthy guinea pigs underwent randomization into the normal control (NC) and LIM group, 1-week (n = 40), 2-week (n = 40), and 6-week (n = 40) subgroups according to the duration of myopia induction. To induce LIM, –6.0 D lenses were placed on the right eyes of guinea pigs in the LIM group, while the left eyes, which received no treatment, served as self-controls. The lenses were kept clean throughout the duration of myopia progression, and the guinea pigs in the NC group received no intervention. In addition, functional verification experiments were conducted using the ROCK inhibitor Y-27632 (SJ-MX1027A, Shandong SparkJade Biotechnology Co., Ltd., Jinan, China). In the Y27632 group of myopic guinea pigs, Y27632 was administered as eye drops to the right eye three times a day at a concentration of 10 mM.

The diopter of the guinea pigs was evaluated before enrollment and after 1 week, 2 weeks, and 6 weeks of myopia induction. Prior to the examination, 10 g$\cdot$L^-1 cyclopentolate hydrochloride eye drops (Alcon, USA) were instilled into the conjunctival sac once every 5 min for a total of 3 times. After the pupil was in a state of complete divergence, a strip retinoscope (Liu Liu Vision Technology Co., Ltd., YZ24, Suzhou, China) was used for binocular diopter detection. The diopter value was obtained from the average of the vertical and horizontal meridians. The ocular axial length was measured by the ophthalmic A-scan ultrasonography (Cinescan, Quantel Medical, France). The instrument parameters were as follows [21]: the propagation velocity of the anterior chamber was 1557 m$\cdot$s^-1, the propagation velocity of the vitreous body was 1540 m$\cdot$s^-1, and the propagation velocity of the lens was 1723 m$\cdot$s^-1. Before the measurements, the guinea pigs in each group were anesthetized with 4 mg$\cdot$mL^-1 Obukaine hydrochloride eye drops (Santen Pharmaceutical Co., Ltd., Japan), and the probe was then gently touched to the cornea to measure the axial length. The average value of 10 continuous measurements represented the final axial length of the guinea pig. All the above measurements of diopter and axial length were conducted by the same inspector under the same operating standards. All experiments were independently repeated at least three times, and all measurement data are presented as means $\pm$ SEM (n = 30).

After 1, 2, and 6 weeks of myopia induction, the guinea pigs in each group were anesthetized by intraperitoneal injection of 3% pentobarbital sodium (30 mg/kg), and the right eyeballs were removed. The ciliary body tissue was then isolated under a microscope. The ciliary body tissue was washed in 1$\times$ PBS rinse solution (CR0014, SparkJade, China), placed in cryogenic vials (NEST Biotech, Wuxi, China), weighed, quickly frozen in liquid nitrogen, and stored at –80 °C. The guinea pigs were euthanized by an additional intraperitoneal injection of pentobarbital sodium (200 mg/kg) to achieve excessive anesthesia.

At the indicated times, the ciliary bodies of guinea pigs were randomly isolated, and total RNA was isolated with a modified SPARKeasy tissue/cell RNA extraction kit (AC0202-B, SparkJade, Jinan, China). Some of these RNA samples were also used for subsequent RT² Profiler™ PCR Array detection. After the transcription of cDNA, the expression levels of the TGF-$\beta$1, RhoA, ROCK1, ROCK2, $\alpha$-SMA, MMP1, Snail1, Slug, Twist1, Twist2, and Zeb1 genes in the ciliary body were detected by RT-qPCR. The primer sequences are listed in Table 1, and the specificity of the RT-qPCR product was confirmed by melting curve analysis. The following PCR cycling program was used: 30 s at 95 °C; 45 cycles of 10 s at 95 °C and 10 s at 62 °C. All experiments were independently repeated at least three times, and these RT-qPCR data are presented as means $\pm$ standard error of the mean (SEM) (n = 6). To quantify relative gene expressions in the ciliary body, transcript levels were normalized to the housekeeping gene Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and calculated via the 2^{-Δ⁢Δ⁢CT} method.

At the indicated times, the ciliary body tissues of the right eye of the guinea pigs in each group were separated (n = 6), and total protein was extracted with radioimmunoprecipitation assay (RIPA) lysis buffer (EA0002, Sparkjade, China) containing phenylmethylsulfonyl fluoride (PMSF) (EA0005, Sparkjade, China). After homogenization and sonication, the lysed tissues were centrifuged at 5000 $\times$g for 5 min, and the supernatants were collected. The concentration of total protein was measured with an enhanced BCA protein concentration assay kit (P0010, Beyotime, Shanghai, China) in a 96-well plate (NEST Biotech., Wuxi, China). The proteins were separated on 10% sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (PAGE) gels (EC1023-B, SparkJade, China) and then transferred to a polyvinylidene fluoride (PVDF) membrane. The PVDF membrane was subsequently blocked with 5% nonfat milk solution for 1.5 h. After blocking, the PVDF membrane was incubated at 4 °C overnight with the primary antibody mixture (dilution ratio = 1:1000, ED0013) (Table 2). The PVDF membrane was then incubated with goat anti-rabbit IgG (H+L) HRP (EF0002, SparkJade, China) diluent (dilution ratio = 1:5000) for 1 hour. Finally, the FUSION FX multifunction imaging system (Vilber Lourmat, Marne-la-Vallée, France) was applied to capture the images and quantified by using the Fusion CAPT software (Vilber Lourmat, France).

Ciliary body tissues of guinea pigs from each group were isolated and placed on ice, followed by protein extraction using ice-cold lysis buffer and subsequent protein quantification. RhoA activity was determined by the G-LISA® RhoA Activation Assay (BK124-L, cytoskeleton, USA). Protein concentrations were normalized to 0.2 mg/mL, and samples were loaded in triplicate along with blank and positive controls onto RhoA-GTP coated plates. After incubating at 4 °C for 30 min with shaking, the plates were washed and incubated with anti-RhoA primary antibody (45 min, room temperature, 400 rpm), followed by a secondary antibody under the same conditions. HRP substrate was added after the final washes, and the reaction was stopped after 15 min at 37 °C. Absorbance was measured at 490 nm using a microplate reader.

To explore the changes in fibrosis-related gene expression in the ciliary body of the LIM guinea pigs, the RT² Profiler™ PCR Array (Qiagen, Germany) for fibrosis was used. Briefly, RNA from the ciliary body in the NC and LIM groups was extracted after myopia induction for 2 weeks, and cDNA was synthesized (AG0302-B, Sparkjade, China). The PCR array assay was performed according to the manufacturer’s instructions. The cycle threshold (CT) values were calculated using the PCR machine’s program, followed by PCR array data analysis.

Differentially expressed genes (DEGs) were defined as those meeting the following criteria: a false discovery rate (FDR) of 0.01, an absolute fold change of $\ge$1.5, and a p-value 0.05. IPA is a web-based software application (http://www.ingenuity.com) that identifies biological pathways and functions associated with biomolecules. After uploading the screened DEGs to the IPA bioinformatics analysis suite (version 1.0, QIAGEN, DUS, Germany), IPA classical signaling pathway analysis, disease and functional analysis, upstream regulatory factor analysis, and downstream disease and functional enrichment analysis were performed.

After 2 weeks of myopia induction, the ocular ciliary body of guinea pigs in the NC and LIM groups was separated under sterile conditions, and total RNA was extracted and purified. The RNA integrity number (RIN) was assessed after purification, and the total RNA content was $\ge$2 µg, the RIN value was $\ge$8, and the OD value/280 ratio was 2.0~2.2 for each sample (n = 3). A single-molecule real-time (SMRT) sequencing platform was used to perform PacBio sequencing for each sample. DESeq2 was used to identify differentially expressed (DE) genes in the NC and LIM groups, employing screening criteria of a fold change between groups $\ge$$\pm$1.5 and an adjusted p value 0.05. Gene annotation enrichment analysis was then performed on the differentially expressed genes using Metascape (https://metascape.org).

After 6 weeks of myopia induction, the eyes from the NC, LIM, and Y27632 groups were separated and immediately placed in EP tubes containing fixative solution for 24 h for routine dehydration, paraffin embedding, and sectioning for Masson’s staining (Servicebio Co., Ltd., Wuhan, China) (n = 3). Briefly, eyeballs were fixed with 4% paraformaldehyde, and the sections were subjected to routine dehydration, paraffin embedding, and dewaxing in water. After overnight incubation with potassium dichromate, ferric hematoxylin, phosphomolybdate, aniline blue, and ponceau acid magenta were used for staining. Finally, the sections were dehydrated and then sealed. With this staining method, collagen fibers appear blue, muscle fibers appear red, and cell nuclei appear blue-black.

After fixing the ciliary body tissue with eye fixative, it was subjected to routine dehydration, paraffin embedding, sectioning, and dewaxing. The tissue sections were then blocked with 5% bovine serum albumin, followed by incubation with $\alpha$-SMA and Col-1 primary antibodies at 4 °C overnight. After washing with PBS, the tissue sections were incubated with a secondary antibody at 37 ℃ for 30 min. Subsequently, the samples were washed with PBS and subjected to DAB staining. Finally, tissue sections were imaged under a light microscope, and the resulting micrographs were quantified using ImageJ software.

After 6 weeks of myopia induction, the eyeballs were separated (n = 3). The same procedure was performed as described for IHC. After antigen retrieval, the sections were washed with PBS (5 min $\times$ 3) and blocked with BSA, which was added dropwise, for 30 min. The primary antibody against TGF-$\beta$1 was added dropwise, followed by incubation at 4 °C overnight. The tissue sections were then incubated with a secondary antibody for 50 min, followed by incubation in the dark with DAPI staining solution for 10 min. The tissue sections were quenched with an autofluorescence quencher for 5 min, sealed with an antifluorescence quencher, and observed by a fluorescence microscope (Nikon, Eclipse, 55i, Tokyo, Japan).

The right eyes of the guinea pigs in the NC and LIM groups (n = 3) were immersed in electron microscope fixation solution (Wuhan Sevier Biotechnology Co., Ltd., Wuhan, China) at 4 °C for 4 h. Next, the ciliary body tissue was cut to approximately 1 mm $\times$ 1 mm $\times$ 1 mm, fixed for 24 h, washed with PBS, and then fixed with 1% osmium acid. The sections were subsequently rinsed with gradient ethanol and dehydrated with acetone. Finally, the tissue sections were observed via TEM (Hitachi, Ltd., HT7700, Japan).

Young’s modulus is a physical quantity that describes the resistance of a solid material to deformation. It is also referred to the tensile modulus. Young’s modulus is the most common type of elastic modulus, and it measures the stiffness of an isotropic elastomer. The mechanical properties of guinea pig ciliary body tissue from each group were evaluated using the Piuma nanoindentation system (OPTICS11, Amsterdam, Netherlands) (n = 3). To minimize the effects of post-mortem tissue autolysis and alterations in biomechanical properties, all nanoindentation tests were completed within 60 min post-enucleation. The entire testing procedure was conducted at a physiological temperature of 37 °C to simulate the in vivo environment, utilizing a stage-top incubator system. To prevent dehydration and subsequent stiffening of the tissue, the ciliary body samples were continuously immersed in pre-warmed PBS throughout the indentation process, ensuring normal hydration and mechanical properties. All indentation tests were performed using a spherical tip with a radius of 27 µm and a cantilever with a stiffness of 0.49 N/m, and the indentation was applied at the center of each piece of ciliary tissue. The elastic modulus of the ciliary tissue was determined, and the data were analyzed using DataViewer v2 (Amsterdam, the Netherlands).

NMT (Younger USA LLC) was used to measure Ca²⁺ fluxes in the ciliary body tissues of guinea pigs in each group, using a 35 mm cell culture dish. Firstly, prepare, install, and calibrate Ca²⁺ selective microelectrodes. Fill the back of the glass microsensor with electrolyte (100 mmol/L CaCl₂), with a length of about 1 cm from the tip; then fill the tip of the glass microsensor with 40 µm of Ca²⁺ liquid ion exchanger (XY-SJ Ca, Younger US); After calibration with 1 mmol/L, 0.5 mmol/L, and 0.05 mmol Ca²⁺ calibration solutions, electrodes with slopes $>$22 mV/decade can be used for subsequent detection. After 6 weeks of myopia induction, the ciliary body tissue of each group of guinea pigs was isolated under sterile conditions (n = 3), immobilized in PBS, and then placed in the test solution. To measure the changes in Ca²⁺ concentration occurring outside the tissue membrane, the electrode movement frequency, or sampling frequency, was set at 0.3 Hz. After recording for 5 min, the raw data microvolt differences ($\mathrm{\Delta }$µV) were imported and converted into the Ca²⁺ flux using JCal version 3.3 (Miami, FL, USA).

To clarify the primary driving role of TGF-$\beta$/RhoA/ROCK pathway activation in myopic ciliary body EMT and fibrosis, we performed functional validation experiments using the ROCK inhibitor Y-27632. After 6 weeks of Y-27632 intervention, the binocular refraction and axial length of myopic guinea pigs were measured. Additionally, Masson staining, IHC, and IF assays were combined to further explore the levels of ciliary body EMT and fibrosis following Y-27632 intervention.

The results are presented as means $\pm$ SEM, and SPSS statistical software (SPSS Version 20.0, Chicago, USA) was used to analyze the data. For comparisons across multiple groups, we first assessed normality and homogeneity of variance. Upon confirmation, one-way ANOVA was applied, followed by Tukey’s post hoc test for all pairwise comparisons. p 0.05 was considered significantly different. Unless otherwise noted, each experiment was repeated three or more times with biologically independent samples.

Before myopia induction, there were no statistically significant differences in diopter values (NC vs. LIM. 3.70 $\pm$ 1.36 D vs. 3.35 $\pm$ 1.03 D) or axial lengths (NC vs. LIM. 8.04 $\pm$ 0.09 mm vs. 8.07 $\pm$ 0.11 mm) between the NC and LIM groups (p $>$ 0.05). After 1, 2, and 6 weeks of myopia induction, the diopter values (NC vs. LIM. 1 w: 2.68 $\pm$ 1.02 D vs. –2.20 $\pm$ 1.49 D; 2 w: 2.52 $\pm$ 1.28 D vs. –2.81 $\pm$ 1.91 D; 6 w: 2.27 $\pm$ 1.05 D vs. –4.93 $\pm$ 1.10 D) of the right eyes in the LIM group were significantly greater than those in the NC group (all p 0.001) (Fig. 1A). Compared with those in the NC group, the differences in diopter values (NC vs. LIM. 1 w: 0.87 $\pm$ 1.22 D vs. –2.60 $\pm$ 0.88 D; 2 w: 0.16 $\pm$ 1.32 D vs. –3.89 $\pm$ 1.77 D; 6 w: 0.47 $\pm$ 0.65 D vs. –5.77 $\pm$ 1.07 D) between the right eyes and the left eyes in the LIM group were significantly greater (all p 0.001) (Fig. 1B). After 1 week, 2 weeks and 6 weeks of lens-induced myopia, the axial lengths (NC vs. LIM. 1 w: 8.17 $\pm$ 0.05 mm vs. 8.23 $\pm$ 0.08 mm; 2 w: 8.18 $\pm$ 0.06 mm vs. 8.29 $\pm$ 0.07 mm; 6 w: 8.40 $\pm$ 0.10 mm vs. 8.72 $\pm$ 0.13 mm) of the right eyes in the LIM group were significantly greater than those in the NC group (all p 0.05) (Fig. 1C), and the differences in axial length (NC vs. LIM. 1 w: 0.00 $\pm$ 0.02 mm vs. 0.05 $\pm$ 0.03 mm; 2 w: 0.02 $\pm$ 0.05 mm vs. 0.09 $\pm$ 0.07 mm; 6 w: 0.03 $\pm$ 0.09 mm vs. 0.14 $\pm$ 0.05 mm) between the right eyes and the left eyes were significantly greater than those in the NC group (all p 0.05) (Fig. 1D). Moreover, during the process of lens-induced myopia, the body weight, food intake and other behavioral activities of guinea pigs in the LIM group did not significantly chang from those in the NC group.

Fig. 1.

Changes in diopter, axial length, TGF-$\beta$/RhoA/ROCK signaling pathway, EMT transcription factors, and fibrosis-related genes in the ciliary body of guinea pigs in each group after 1-, 2-, and 6-week myopia induction. (A) The diopter of the right eye in the LIM group was significantly higher than that in the NC group. (B) The diopter difference between the right eye and the left eye in the LIM group was significantly increased compared with the NC group. (C) The axial length of the right eye in the LIM group increased compared with the NC group. (D) The difference between the axial lengths of the right eye and the left eye in the LIM group was significant. (E) The expression of TGF-$\beta$1 mRNA. (F) The expression of RhoA mRNA. (G) The expression of ROCK1 mRNA. (H) The expression of ROCK2 mRNA. (I) The expression of $\alpha$-SMA mRNA. (J) The expression of MMP1 mRNA. (K) The expression of Snail1 mRNA. (L) The expression of Slug mRNA. (M) The expression of Twist1 mRNA. (N) The expression of Twist2 mRNA. (O) The expression of Zeb1 mRNA. (P) The diopter changes of the right eye in the LIM group in the ultra-early stage of lens-induced myopia. (Q) The axial length changes of the right eye in the LIM group in the ultra-early stage of lens-induced myopia. (R) The expression of Snail1 mRNA in the ultra-early stage of lens-induced myopia. (S) The expression of Slug mRNA in the ultra-early stage of lens-induced myopia. (T) The expression of Twist1 mRNA in the ultra-early stage of lens-induced myopia. (U) The expression of Zeb1 mRNA in the ultra-early stage of lens-induced myopia. All RT-qPCR data are presented as mean $\pm$ standard error of the mean (SEM) (n = 6). *p 0.05, **p 0.01, and ***p 0.001. TGF, transforming growth factor; RhoA, Ras homolog family member A; ROCK, Rho-associated protein kinase; $\alpha$-SMA, $\alpha$-smooth muscle actin; MMP, matrix metalloproteinase; LIM, lens-induced myopia; NC, normal control.

After 1, 2, and 6 weeks of myopia induction, the mRNA expression levels of TGF-$\beta$1 (Fig. 1E), RhoA (Fig. 1F), ROCK1 (Fig. 1G), ROCK2 (Fig. 1H),$\alpha$-SMA (Fig. 1I), and MMP1 (Fig. 1J) in the ciliary body of the LIM group were significantly greater than those in the NC group (all p 0.01), and these TGF-$\beta$/RhoA/ROCK signaling pathway components and fibrosis-related genes increased more significantly after 6 weeks of myopia induction. Furthermore, the mRNA expression levels of EMT transcription factors such as Snail1 (Fig. 1K), Slug (Fig. 1L), Twist1 (Fig. 1M), Twist2 (Fig. 1N), and Zeb1 (Fig. 1O) in the ciliary body of the LIM group were significantly higher than those in the NC group after 1-, and 2-week myopia induction (all p 0.01), confirming that EMT occurred in the ciliary body at the early stage of myopia. However, to clarify whether EMT-mediated fibrosis is the initiating event or merely a consequence of prior functional alterations, we dynamically monitored EMT transcription factors, refractive error, and axial length parameters at ultra-early time points (1, 3, and 5 days). This aimed to evaluate the relationship between the temporal sequence of EMT and functional changes with earlier times. Results showed that on day 1 after lens induction, there were no significant differences in the refractive error or axial length of the right eye between the LIM group and the NC group (Fig. 1P,Q), nor were there notable changes in the mRNA levels of EMT-related transcription factors (Fig. 1R–U). On days 3 and 5 after lens induction, the refractive error and axial length of the right eye in the LIM group increased significantly (all p 0.01), while the mRNA levels of EMT-related transcription factors remained unchanged (Fig. 1R–U). These findings indicate that functional changes, such as increases in biological parameters of refractive error and axial length, occur earlier than alterations in EMT transcription factor levels during the ultra-early stage of lens-induced myopia. This suggests that EMT-mediated fibrosis is not an initiating event but merely a consequence of prior functional modifications.

As shown in Fig. 2A, after 1, 2, and 6 weeks of myopia induction, the protein levels of TGF-$\beta$1 (Fig. 2B), ROCK1 (Fig. 2D), and ROCK2 (Fig. 2E) in the ciliary body of LIM eyes were significantly greater than in the NC group (all p 0.05). In addition, the protein expression of RhoA (Fig. 2C), $\alpha$-SMA (Fig. 2F), and MMP1 (Fig. 2G) in the ciliary body of LIM eyes was significantly greater in the LIM group than in the NC group (all p 0.01), and the increases were especially significant at 6 weeks after myopia induction (all p 0.01). The protein expression of N-cadherin (Fig. 2H,J) in the ciliary body of guinea pig eyes in the LIM group after inducing myopia for 1, 2, and 6 weeks was significantly higher than that in the NC group (all p 0.05). After inducing myopia for 2 and 6 weeks, the expression of E-cadherin (Fig. 2H,I) protein in the ciliary body of guinea pig eyes in the LIM group increased (all p 0.05), and the expression was particularly significant at 6 weeks after inducing myopia (all p 0.05). Moreover, the results showed that after inducing myopia for 1, 2, and 6 weeks, the activity of RhoA in the ciliary body of guinea pigs in the LIM group was higher than that in the NC group (Fig. 2K) (all p 0.05), and it was particularly significant at 6 weeks, suggesting that the occurrence and development of myopia are closely related to the activation of the RhoA signaling pathway.

Fig. 2.

The protein levels related to the TGF-$\beta$/RhoA/ROCK signaling pathway and fibrosis in the ciliary body. (A) TGF-$\beta$/RhoA/ROCK signaling pathway and fibrosis related protein expression. (B–G) Histogram of optical density analysis in TGF-$\beta$1, RhoA, ROCK1, ROCK2, $\alpha$-SMA and MMP-1 proteins. (H) Expression of E-cadherin and N-cadherin proteins. (I) Histogram of optical density analysis in E-cadherin protein. (J) Histogram of optical density analysis in N-cadherin protein. (K) G-LISA detection of RhoA activity. All data are presented as mean $\pm$ SEM (n = 6). Compared with the NC group, *p 0.05, **p 0.01, and ***p 0.001. Compared with the LIM group in week 1, ^#p 0.05, ^#⁢#p 0.01, and ^#⁢#⁢#p 0.001.

Fig. 3 shows that there were 42 DEGs in the LIM group (vs. NC), including 35 upregulated DEGs (red) and 7 downregulated DEGs (blue) (Fig. 3A). Further analysis of the DEGs revealed that the upregulated genes accounted for 41.67% of the 84 gene arrays, and 31.43% of the upregulated genes had differential multiples $\ge$2.0, indicating that many fibrosis-related mechanisms were activated in the ciliary body of guinea pigs with myopia (Fig. 3B). Moreover, cluster analysis of the expression degree of DEGs in all the groups revealed upregulated genes related to fibrosis, which involved mainly Tgfbr2, Col3a1, Col1a2, and Hgf, as well as matrix metalloproteinase (MMP) family genes (Mmp2, Mmp3, Mmp9, and Mmp14). These findings suggested that the pathological process of myopia is associated with the activation of the TGF-$\beta$ signaling pathway and the involvement of tissue repair processes such as fibrosis (Fig. 3C).

Fig. 3.

Changes in 84 gene levels related to fibrosis and Ingenuity pathway analysis of DEGs in the ciliary body fibrosis PCR array of the LIM group in the ciliary body tissue of each group after 2-week myopia induction (n = 3). (A) Volcano map of 84 fibrosis related genes, compared with the NC group, among the DEGs in the LIM group with a fold difference $\ge$$\pm$1.5 and p 0.05. (B) Pie chart of 84 fibrosis-related genes in different categories. (C) Heat maps of 42 DEGs detected by 84 gene array of fibrosis, the gene clustering was enrichment analyzed based on the similarity of gene expression between samples. (D) IPA classical signaling pathway analysis. (E) Disease and functional analysis. (F) Upstream regulatory factors and downstream disease and function enrichment analysis. DEGs, Differentially expressed genes; IPA, Ingenuity pathway analysis.

IPA revealed that the pathways associated with the DEGs were activated in the LIM group compared with those in the NC group (Fig. 3D), and these pathways involved mainly the osteoarthritis pathway, the hepatic fibrosis signaling pathway, the regulation of EMT by the growth factor pathway, and the TGF-$\beta$ signaling pathway. Disease and function analyses of the DEGs revealed that the differential genes in the LIM group (vs. the NC group) were related mainly to cellular growth and proliferation, cellular movement, cellular development, and immune cell trafficking. In addition, the DEGs in the LIM group (vs. the NC group) affected organizational development, organizational injury and abnormalities, connective tissue development, and function (Fig. 3E). Thus, during the development of myopia, cell metabolism and homeostasis may be significantly affected, and the development and function of tissues and organs in the body may be significantly altered, eventually leading to tissue damage and abnormalities (Fig. 3E). Moreover, the upregulation of MMP8 and HGF may lead to the inhibition of ciliary body fibrosis in guinea pigs with myopia (Fig. 3F).

In total, 139 DEGs were identified in the ciliary muscle tissue of guinea pigs in the NC group and LIM group, including 29 upregulated genes and 120 downregulated genes. The cell type dot plot revealed that the DEGs were associated mainly with fetal retina bipolar cells, fetal retina photoreceptor cells, adult olfactory neuroepithelial stromal fibroblasts and fetal eye photoreceptor cells (Fig. 4A), and atypical scarring of the skin was enriched in the analysis of the disease gene network (Fig. 4B). GO enrichment analysis revealed that these DEGs were related mainly to the biological process of protein localization to the photoreceptor outer segment (Fig. 4C) and to the myometrial relaxation and contraction pathways (Fig. 4D), which provides evidence for the observed dysfunction of ciliary body elasticity in myopia. In addition, RHO GTPase activation of IQGAPs, adherens junction interactions, and VEGFR2-mediated vascular permeability were enriched in the Reactome gene sets (Fig. 4E). IQGAPs are key effector proteins and intersections of RhoA and Rac1. Enrichment analysis suggests significant changes in the activity of the Rho GTPase family (including RhoA) in the myopic ciliary body. GO enrichment analysis of the transcription factor targets of the DEGs revealed associations with the following terms: skeletal muscle cell differentiation, skeletal muscle tissue development, and skeletal muscle organ development (Fig. 4F); MAPK targets/nuclear events mediated by MAP kinases and MAP kinase activation (Fig. 4G); and the effect of MFAP5 on the permeability and motility of endothelial cells via cytoskeleton rearrangement (Fig. 4H). Notably, the COL3A1 and TGF$\beta$1 DEGs were involved in visual perception, the formation of annular gap junctions, the Beta-catenin-independent WNT signaling pathway, and other biological functions according to the DEG distribution map (Fig. 4I), suggesting the activation of the TGF-$\beta$/RhoA signaling pathway and the occurrence of fibrosis in myopic body tissues. These findings indicated that myometrial relaxation and contraction pathways, as well as adhesion junction interactions, are involved in the occurrence of myopia, which is strongly correlated with ciliary body fibrosis.

Fig. 4.

GO enrichment analysis of the differentially expressed mRNAs and Transcription factor targets. (A) Celltype dotplot of differentially expressed genes (DEGs). (B) DisGeNet barplot of DEGs. (C) Biological process (BP) dotplot of DEGs. (D) WikiPathways dotplot of DEGs. (E) Reactome Gene set dotplot of DEGs. (F) Biological process (BP) dotplot of Transcription factor targets. (G) WikiPathways dotplot of Transcription factor targets. (H) Reactome gene set dotplot of Transcription factor. (I) DEGs distribution map.

Compared with that in the NC group, the degree of fibrosis (as indicated by blue color) (NC vs. LIM. 0.24763159 $\pm$ 0.013337% vs. 0.408256574 $\pm$ 0.013965%) in the ciliary body of guinea pigs in the LIM group was aggravated after 6 weeks of myopia induction (Fig. 5A,B). In addition, immunohistochemistry was utilized to evaluate the levels of $\alpha$-SMA and Col-1 in the ciliary body tissues of guinea pigs with myopia. The expression of $\alpha$-SMA (NC vs. LIM. 0.1734483 $\pm$ 0.0431506% vs. 0.3568124 $\pm$ 0.0344084%) and Col-1 (NC vs. LIM. 0.24763159 $\pm$ 0.013337% vs. 0.408256574 $\pm$ 0.013965%) in the ciliary body of guinea pigs in the 6-week LIM group was greater than that in the NC group (Fig. 5C,D), suggesting that ciliary body fibrosis and activation of fibroblasts occurred in the process of myopia.

Fig. 5.

The fibrotic progression of the ciliary body tissue detected by Masson Staining, immunohistochemistry, and immunofluorescence after 6-week myopia induction. (A) Masson Staining (400$\times$), Bar = 20 µm. (B) Histogram of relative collagen content analysis. Compared with the NC group, ***p 0.001. (C) Immunohistochemistry (400$\times$), Bar = 20 µm. (D) Histogram of optical density analysis in $\alpha$-SMA and Col-1 proteins. Compared with the NC group, *p 0.05 and **p 0.01. (E) The $\alpha$-SMA levels in the ciliary body (400$\times$), Bar = 20 µm. (F) The Col-1 levels in the ciliary body (400$\times$), Bar = 20 µm. (G) Histogram of optical density analysis in $\alpha$-SMA and Col-1 proteins. All data are presented as mean $\pm$ SEM (n = 3). Compared with the NC group, **p 0.01, and ***p 0.001.

In Fig. 5, the nuclei stained with DAPI appear blue, and the corresponding fluorescein-labeled positive expression of $\alpha$-SMA (Fig. 5E) and Col-1 (Fig. 5F) is shown in red. The expression of $\alpha$-SMA (NC vs. LIM. 0.1691442 $\pm$ 0.0258369% vs. 0.616374 $\pm$ 0.0341538%) and Col-1 (NC vs. LIM. 0.1174157 $\pm$ 0.0608709% vs. 0.7198452 $\pm$ 0.0560648%) in the ciliary body tissue of the LIM group was greater than that in the NC group (Fig. 5G), indicating the myofibroblast activation and ciliary body fibrosis in the pathological process of myopia.

The structural characteristics of the ciliary body in each group were evaluated by TEM. The epithelial cells in the NC group had a complete cell membrane, a regular morphology, tight cell connections, and fewer fibroblasts (Fig. 6A). In the LIM group, the epithelial cells lacked a regular morphology, and the cells were slender, with fibroblast-like characteristics, indicating the occurrence of ciliary body EMT and fibrosis in myopia (Fig. 6B).

Fig. 6.

The ciliary body of the NC group and LIM group was observed by TEM, Piuma nanoindentation system, and Non-invasive Micro-Test Technology (NMT) after 6-week myopia induction (10000$\mathrm{\times }$). All data are presented as mean $\pm$ SEM (n = 9). (A) The ciliary body of the NC group was observed by TEM. Bar = 2 μm. (B) The ciliary body of the LIM group was observed by TEM (n = 3). $\to$: Epithelial Cells, $△$: Fibroblasts. Bar = 2 μm. (C,D) Displacement-time curve. (E,F) The load-indentation curve of the nanoindentation test depicts the dynamic change of load and indentation from load to hold to unloading when subjected to external force. (G) Young’s modulus statistic plot (n = 3). Compared with the NC group, ***p 0.001. (H) NMT operation chart, (I) Ca²⁺ flux, and (J) Analysis of Ca²⁺ flux (n = 3). Compared with the NC group, ***p 0.001. TEM, Transmission electron microscopy.

Young’s modulus is the ratio of stress ($\sigma$) to strain ($ϵ$) in the linear elastic deformation stage of a structured material. Young’s modulus is the core quantitative index of the elastic properties of a structured material, and it directly reflects the resistance of a structured material to external forces in the elastic deformation stage. After 6 weeks of myopia induction, the Young’s modulus of the ciliary body tissue in the guinea pigs in the NC and LIM groups was measured. Compared with the NC group, the LIM group presented an increase in the Young’s modulus of the ciliary body tissue in guinea pigs (Fig. 6C–G). A higher Young’s modulus indicates a stiffer structure and therefore a smaller elastic deformation for a given applied load. The increase in elastic modulus and decrease in elasticity of the ciliary body tissue in the LIM group guinea pigs further confirmed the occurrence of fibrosis in the ciliary body of guinea pigs with myopia.

After 6 weeks of myopia induction, the Ca²⁺ levels in the ciliary body tissue of guinea pigs in the NC and LIM groups were measured via NMT (Fig. 6H). Abnormal intracellular Ca²⁺ regulation is involved in the occurrence and development of tissue fibrosis, and an increase in the intracellular Ca²⁺ level can promote tissue fibrosis. Compared with that in the NC group, the level of Ca²⁺ inflow in the ciliary body tissue of guinea pigs in the LIM group was increased, which indicated that the occurrence of myopia was related to the aggravation of ciliary body fibrosis (Fig. 6I,J).

The results showed that after Y-27632 intervention, both the refractive error of the right eye and the interocular refractive error difference in guinea pigs were lower than those in the lens-induced myopia (LIM) group (Fig. 7A,B) (all p 0.05), and the axial length as well as the interocular axial length difference were also significantly improved (Fig. 7C,D) (p 0.05). Masson staining revealed that compared with the LIM group, the degree of ciliary body fibrosis in guinea pigs was reduced after Y-27632 intervention (Fig. 7E,F). Moreover, IHC and IF were used to evaluate the levels of $\alpha$-SMA and Col-1 in the ciliary body tissues of myopic guinea pigs after ROCK inhibition. The expressions of $\alpha$-SMA and Col-1 in the ciliary body of the group were lower than those in the LIM group (Fig. 7G–J), suggesting that ROCK inhibition can ameliorate the pathological levels of EMT and fibrosis in the ciliary body of myopic guinea pigs.

Fig. 7.

The ocular biological parameters, ciliary body EMT and fibrosis levels of myopic guinea pigs after inhibiting ROCK for 6 weeks. (A) Refractive error of the right eye in guinea pigs from the LIM group and Y-27632 group. (B) Interocular difference in refractive error between the right and left eyes of individual guinea pigs in the LIM group and Y-27632 group. (C) Axial length of the right eye in guinea pigs from the LIM group and Y-27632 group. (D) Interocular difference in axial length between the right and left eyes of individual guinea pigs in the LIM group and Y-27632 group. (E) Masson Staining (400$\times$), Bar = 20 µm. (F) Histogram of relative collagen content analysis. (G) Immunohistochemistry (400$\times$), Bar = 20 µm. (H) Histogram of optical density analysis in $\alpha$-SMA and Col-1 proteins. (I) The Col-1 levels in the ciliary body (400$\times$), Bar = 20 µm. (J) Histogram of optical density analysis in $\alpha$-SMA and Col-1 proteins. All data are presented as mean $\pm$ SEM (n = 3). Compared with the LIM group, *p 0.05, **p 0.01, and ***p 0.001.