Evidence for M₂ Muscarinic Receptor Antagonist Delay of Myopia Development Through Activation of Kir3.4 Channel in the Retina of Guinea Pigs

Abstract

Purpose: The purpose of this study was to investigate the association between muscarinic receptor M₂ and potassium channel Kir3.4 encoded by gene KCNJ5, as well as their role in guinea pigs with form deprivation myopia (FDM).

Methods: One hundred sixty-five 3-week-old guinea pigs were randomly assigned to the following groups: normal control (NC), self-control (SC), form deprivation (FD), lentiviral vector (FD + Vector), KCNJ5 overexpression lentivirus (FD + KCNJ5-OE), vehicle control (FD + DMSO), M₂ receptor antagonist (FD + AF-DX 116), and M₂ receptor agonist (FD + LY2119620). The association between M₂ receptors and retinal potassium channels and effects of retinal K⁺ concentration on myopia development were investigated by constructing a lentiviral KCNJ5 overexpression and M₂ receptor intervention model. Immunohistochemistry and molecular assays were conducted to measure the distribution and expression of Kir3.4-related mRNA and protein in the retina. TUNEL was used to observe the drug toxicity response on the retina.

Results: The FD group had higher myopic degree (all P < 0.001) and lower expression levels of Kir3.4 than the NC group (P = 0.008). The FD + KCNJ5-OE group exhibited upregulated Kir3.4 protein expression (P < 0.001), but a significant decrease in myopia degree and K⁺ concentration (all P < 0.001) compared with the FD + Vector group. The FD + AF-DX 116 group exhibited lower myopic degree, K⁺ concentration (all P < 0.05), and higher Kir3.4 protein expression (P < 0.001), as well as the FD + LY2119620 group exhibited significantly upregulated myopia degree and K⁺ concentration (all P < 0.001) compared with the FD + DMSO group.

Conclusions: This study is the first to explore the muscarinic receptor-potassium channel connection and its implications in the development of myopia. The M₂ receptor may be involved in the development of myopia by regulating retinal Kir3.4 channel and K⁺ homeostasis.

Myopia is one of the major contributors to visual impairment worldwide. As refractive error and axial length (AL) increases, complications such as retinal and choroidal degeneration or atrophy usually occur and progress, even leading to irreversible vision damage.¹ Currently, topical atropine sulfate is the primary drug used clinically for the management of myopia in children.^2–4 Pharmacologically, atropine functions as a nonselective antagonist of muscarinic acetylcholine receptors (MRs), exhibiting an affinity for all five subtypes of MRs (M₁R–M₅R). These receptors, expressed in the ocular cornea, the sclera, and the retina,^5–7 are involved in eyeball growth and cell-to-cell signaling in the retina.⁸ Barathi et al.⁹ found that mice lacking the M₂ receptor gene exhibited reduced susceptibility to lens-induced myopia, and proposed that the M₂ receptor may be involved in myopia development through ion channels. The above research tips that muscarinic receptor M₂ plays a significant role in myopia protection. The M₂ receptor-selective antagonists can avoid the side effects, such as pupil dilation and photophobia caused by the M₃ receptor in atropine. Elucidating the mechanism of M₂ receptor involvement in myopia prevention and control mechanism is expected to bring new targets for the clinical treatment of myopia. 

MRs belong to the superfamily of guanosine triphosphate-binding protein-coupled receptors, which regulate the activity of various effectors such as ion channels and enzymes, by coupling with G proteins for intracellular signaling.¹⁰ Previous studies have demonstrated that MRs have to influence neural excitability transfer by interacting with endogenous ion channels in order to modify information transfer.¹¹ The acetylcholine-sensitive potassium channel Kir3.4, encoded by the potassium channel gene KCNJ5, belongs to the G protein-coupled inward rectifier potassium channel (GIRK or Kir3). GIRK has been shown to couple with the M₂ receptor to maintain the membrane resting potential and regulate neuronal excitability.¹² The membrane resting potential is closely related to the potassium ion (K⁺) homeostatic potential, and an imbalance of K⁺ homeostasis in ocular tissues is a significant indicator of myopia progression.^13,14 Therefore, it is speculated that the M₂ receptor influences the development of myopia by regulating Kir3.4 potassium channels and potassium ion concentration levels. 

Therefore, the present study aimed to investigate the mechanism that muscarinic receptors influenced myopia development by detecting the effects of M₂ receptor intervention and KCNJ5 overexpression on Kir3.4 potassium channels and K⁺ concentration and ocular bioparameters, and investigate whether the M₂ muscarinic receptor antagonists delay myopia progression by regulating the Kir3.4 potassium channel and retinal K⁺ concentration in guinea pigs. 

Materials and Methods

Animals and Grouping

A total of 165 healthy 3-week-old SPF-grade tricolor male guinea pigs were provided by the Sichuan Provincial Specialized Committee for Laboratory Animals Farm (License No. SYXK [Sichuan] 2023-0076), weighing 150 to 180 g. All guinea pigs were housed at 25 ± 2°C with free access to food and water, and subjected to a 12-hour light/ dark cycle (light hours = 8:00–20:00) with a light intensity of approximately 300 to 500 lux (lx). The guinea pigs were fed and handled in full compliance with the Statement for Animal Research in Vision and Ophthalmology, and the study protocol was reviewed and approved by the Ethics Committee of North Sichuan Medical College (No. 2024019). 

Using a randomized numeric table method, the guinea pigs were divided into the normal control (NC) group (n = 15) and an experimental group in the first part. The NC group of guinea pigs had both eyes left untreated. The left eye of the experimental group without any treatment was set as the self-control (SC) group (n = 15), and the right eye was covered with a translucent latex balloon (allows light to enter but removes the sense of clarity to reduce image contrast and spatial frequency) for 4 weeks to induce myopia as the form deprivation (FD; n = 15) group.¹⁵ In addition, the nose, mouth, and ears were exposed. The tightness and coverage of the facemask were checked every day. In the second part, overexpressing lentivirus were constructed and guinea pigs were assigned to following groups: NC (n = 15), FD (n = 15), lentiviral vector (FD + Vector; n = 15), and KCNJ5 overexpression lentivirus (FD + KCNJ5-OE; n = 15). The lentiviral vector was an empty vector and did not contain any positively expressed virus to exclude the potential effects of the viral vectors on KCNJ5 expression. In the third part, guinea pigs were grouped according to different drug treatments: NC (n = 15), FD (n = 15), vehicle control (FD + DMSO; n = 15), M₂ receptor antagonist (FD + AF-DX 116; n = 15), and M₂ receptor agonist (FD + LY2119620; n = 15). The DMSO is the solvent of the intervention drug and is used to dissolve the drug for injection. 

The guinea pig was lightly anesthetized with isoflurane and disinfected around the eyes with iodophor, then an intravitreal injection was administered. The lentivirus (Songon, China) group only received one injection (1.0*10⁹ TU/mL, 5 µL) on day 1 of form deprivation myopia (FDM) modeling. The M₂ receptor intervention group was injected every 4 days starting on day 1 of modeling with an injection volume of 10 µL, 10% DMSO, 200 µM AF-DX 116, and 100 µM LY2119620 (GlpBio, USA). DMSO only and drug only effects are assumed negligible based on prior studies. 

After 4 weeks, the guinea pigs were deeply anesthetized with excess isoflurane until the corneal and pain reflexes disappeared. The eyeballs were immediately removed. Some were fixed in 4% paraformaldehyde (Solarbio, China) for embedded sections and the remaining were rapidly excised on ice to remove the cornea, lens, and vitreous, then the retina was isolated to be frozen in liquid nitrogen and stored at –80°C for subsequent experiments. 

Biological Measurements

Refractive power and AL measurements were performed before (0 week) and 1, 2, 3, and 4 weeks after the experiments. The pupils were fully dilated with compound tropicamide eye drops (Santen, Japan), and the refraction was measured in the dark room using a retinoscopy (66 Vision Technology Co., China), during which the guinea pig’s line of sight was fixed and the microscope was aligned with its pupil. The refractive power is examined by the same experienced optometrist using a retinal ophthalmoscope, measured five times, and averaged after removing the maximum and minimum values. Then, the refractive power was adjusted to neutralization and converted to the spherical equivalent (SE; spherical power + 1/2 cylindrical power) for recording. Oxybutynin hydrochloride eye drops (Santen) were used to anesthetize the corneal surface, and the measurements were started when the probe touched the corneal surface without inducing a blinking reflex. The AL was measured five times by the same experienced special inspection technician using an A-model ultrasonic instrument (Quantel Medical, France), repeated until consistent (<10 µm variation), and the maximum and minimum values were removed and averaged. 

Hematoxylin and Eosin Staining

The paraffin sections of 5 µm thickness were prepared. Staining was performed according to the standard instructions of hematoxylin and eosin staining kit (Biosharp, China). Paraffin sections were dewaxed and hydrated, then soaked in hematoxylin for 3 minutes. The sections were treated with 1% hydrochloric acid alcohol for 2 seconds. Eosin staining was performed for 1 minute. The sections were dehydrated and clear. Finally, the sections were sealed with neutral resin for observation under a light microscope (Leica, France). Retinal and choroidal thickness was measured using a measuring tool ruler under 40× magnification. 

Immunohistochemical Staining

The paraffin sections of 3 µm thickness were prepared for staining according to the immunohistochemistry kit (SA1027; Boster, China). Sections were deparaffinized and hydrated for antigen repair for 20 minutes, incubated in 3% hydrogen peroxide-methanol solution for 15 minutes, and incubated at 37°C for 1 hour. Primary antibody KCNJ5 (1:100; Invitrogen, USA) was added and incubated at 4°C overnight. The anti-rabbit secondary antibody (SA1027; Boster, China) was incubated for 1 hour at room temperature. Immunoreactivity was subsequently tested with DAB reagent (ZSGB-BIO, China). Hematoxylin was used to re-stain the sections and dehydrated for transparency. Finally, the sections were sealed with neutral resin and observed under a light microscope (Leica). 

Measurement of Potassium Ion Concentration

Fresh retinal tissue from guinea pigs was extracted according to the instructions of the K⁺ assay kit (C001-2-1; Nanjing Jiancheng Bioengineering Institute, China), and homogenized with pre-cooled sterile deionized water (g/mL = 1:9) and then centrifuged (10,000 × g for 10 minutes at 4°C). The supernatant was placed on ice for measurement. The protein concentration of the supernatant was determined by the bicinchoninic acid (BCA) method. The supernatant was then mixed with the working solution in the kit and left at room temperature for 5 minutes. The absorbance value was measured at 440 nm using an enzyme marker (Molecular Device, USA). Finally, the potassium ion concentration was calculated according to the formula. 

Quantitative RT-PCR

Total retinal RNA was extracted using the Trizol (Beyotime, China) method, and cDNA was prepared using the PrimeScript RT Master Mix (Takara Bio, Japan) reverse transcription kit. The designed primers were amplified using SYBR Premix Ex Taq II (Takara) on a Roche Light Cycler 480 instrument. The amplification was performed under the following cycling conditions: 95°C for 30 seconds, 95°C for 5 seconds, 60°C for 30 seconds, and 97°C for 1 second, for a total of 40 cycles. The sequences were as follows: KCNJ5 forward primer 5′-CCTACATCCGAGGCGATCTG-3′ and reverse primer 5′-ATGAAGGCGTTGACGATGGA-3′, CHRM2 forward primer 5′-GGCAGGCATGATGATTGCAG-3′ and reverse primer 5′-TGCCAGAAGAGAATGGCTGG-3′, GAPDH forward primer 5′-GGTATTCCTTCTTCCCGTGC-3′, and reverse primer 5′-CCAAATCCGTTCACTCCGA-3′. At the end of amplification, the relative expression of target genes was calculated by the 2^−∆∆Ct method using GAPDH as an internal reference. 

Western Blot Analysis

RIPA lysis buffer (Solarbio, China) was added to the retinas and sonicated (Qsonica, USA). The supernatant was collected after centrifugation at 12,000 × g for 15 minutes, and the protein concentration was determined using the BCA kit (Beyotime). Samples were mixed with 5× Protein Sampling Buffer (Beyotime) and placed on a metal bath at 100°C for 15 minutes. The 12% One-Step PAGE gels were used for constant pressure electrophoresis for 1.2 hours, and the membranes were transferred by wet-turning for 1.1 hours. The membranes were incubated with 5% skimmed milk for 2 hours at room temperature, and primary antibodies (anti-KCNJ5, 1:650, PA5-119662, Invitrogen; anti-CHRM2, 1:4000, AB109226, Abcam; anti-GAPDH, 1:5,000, and ET1601-4; Huabio) were added and incubated overnight at 4°C. Afterward, the membrane was incubated with secondary antibody (1:6000; Boster) for 1 hour at room temperature. Finally, the strips were exposed using ECL developer solution (Biosharp) and the optical density was quantified using Image J software. 

Retinal Pigment Epithelium Primary Cell Isolation and Culture

The eyeballs were removed and rinsed in sterile PBS solution, and then placed in triple antibiotic and immersed 3 times (10 minutes/time). The anterior segment eye tissue was removed by cutting along the corneoscleral margin. The posterior segment eye cups were washed in pre-cooled Hanks’ balanced salt solution (HBSS; Viva Cell, China) and digested in 0.25% trypsin digestion solution (Viva Cell, China) for 30 minutes at 37°C in an incubator. Digestion was terminated with fresh DMEM/F12 (Viva Cell, China) medium containing 20% fetal bovine serum. The cells were centrifuged at 1000 rpm/min for 10 minutes, and the cell precipitate was resuspended in DMEM/F12 medium. The medium was changed every 2 to 3 days and the cells were passaged at 80% to 90% fusion. 

Lentiviral Transfection

When the fusion rate of the cells was approximately 40% after wall attachment, KCNJ5 overexpression lentivirus and lentiviral vector were added for viral transfection, respectively (multiplicity of infection [MOI] = 10), and the infection efficiency reached the peak after 72 hours. Cultivation with culture medium containing puromycin (2 µg/mL) and the setup of a blank cell control well at the same time. The blank cell death rate in the control well was observed after 48 hours. The puromycin concentration was considered appropriate if the cell death rate exceeds 90%. Then, we obtained the stably virus-transfected cell lines after several times of screening. 

TUNEL Apoptosis Staining

Staining was performed according to the standard operating instructions of TUNEL staining kit (11684795910, In Situ Cell Death Detection Kit, Fluorescein, Switzerland). Paraffin sections were dewaxed and hydrated. The tissues were treated with Proteinase K working solution for 30 minutes. Tissues were incubated with TdT and fluorescein-labeled dUTP solution for 1 hour at 37°C, avoiding light. DAPI staining solution was incubated for 10 minutes at 37°C, avoiding light. Then slices were blocked with an anti-fluorescence quencher. The slices were observed under a fluorescence microscope and photographed. Image J software was used to measure the optical density value of retina positive cells. 

Statistical analyses

The data are presented as the mean ± SD and analyzed by SPSS version 27.0 software (SPSS, USA), and GraphPad Prism version 9.5 (GraphPad, USA) was utilized for graphing. A 2-way ANOVA followed by the Bonferroni multiple comparisons test was used to assess the biological parameters of each group at various time points. Paired t-tests were used to assess the significance between the FD and the SC eye. Molecular level differences among groups were examined using 1-way analysis of variance. All results shown represent at least 3 independent experiments and P < 0.05 was considered statistically significant. 

Results

KCNJ5 Gene Involved in FDM Development

Prior to myopia induction, most of the refractive states of the 3-week-old guinea pigs were hyperopic, and there were no significant differences in SE and AL (P = 0.879 and P = 0.915, respectively). After 1 week of treatment, the FD group began to show myopic drift. Four weeks of myopia induction resulted in a significant increase in refractive error (95% confidence interval [CI] = –3.54 to –3.00) and AL (95% CI = 8.27–8.40, all P < 0.001; Table 1; Figs. 1A, 1B). 

Table 1.

 

Changes in SE and AL in Different Groups (\(\overline{{\rm{X}}}\) ± s)

 

Figure 1.

 

Changes in ocular bioparameters and Kir3.4 expression after form deprivation. (A) Changes in refractive state at 0, 1, 2, 3, and 4 weeks after form deprivation. (B) Changes in AL at 0, 1, 2, 3, and 4 weeks after form deprivation. (C) Hematoxylin and eosin staining results of retina in each group after 4 weeks. (D) Statistical results of total retinal thickness. (E) Hematoxylin and eosin staining results of choroid in each group after 4 weeks. (F) Statistical results of total choroidal thickness. (G) Immunohistochemical staining of retina for Kir3.4 protein in each group after 4 weeks. (H) Statistical results of mean optical density values of immunohistochemical images. (I) The expression of Kir3.4 at mRNA level in each group after 4 weeks. (J) Western blot electrophoretic bands of Kir3.4 expressed after 4 weeks. (K) The expression of Kir3.4 at protein level in each group after 4 weeks. GCL, ganglion cell layer; IPL, inner plexiform layer; INL, inner nuclear layer; OPL, outer plexiform layer; ONL, outer nuclear layer; IS/OS, inner and outer segments of photoreceptor; RPE, pigment epithelial layer. Scale bar = 50 µm. *P < 0.05, **P < 0.01, ***P < 0.001.

It was observed by hematoxylin and eosin stain that the tissues of the NC and SC groups were relatively intact and aligned, whereas the tissues of the FD group showed fracture structures and disordered alignment (Fig. 1C). In addition, the thickness of the retina and choroid in the FD group was significantly thinner than that in the NC and SC groups (retina = P = 0.001 and P = 0.01; and choroid = P = 0.008 and P < 0.001; see Figs. 1D–F). The above results suggest that the model of FDM was successfully constructed. Immunohistochemistry showed that the Kir3.4 protein was mainly distributed in the ganglion cell layer, inner nuclear layer, outer nuclear layer, and the inner and outer segments of the photoreceptor and the retinal pigment epithelium (RPE), with the strongest expression detected in the RPE and outer segments of the photoreceptor (Fig. 1G). Compared with the NC and SC groups at the corresponding time point, the Kir3.4 optical density values in the FD group showed a trend of downregulation, and the differences were statistically significant (P = 0.009 and P = 0.041, respectively; Fig. 1H). 

After 4 weeks, the expression of Kir3.4 mRNA was detected in each group, the FD group was significantly lower than the NC and SC groups (P = 0.008 and P = 0.016, respectively; Fig. 1I). The expression of Kir3.4 protein in the guinea pig retina was further examined by Western blot, and the Kir3.4 protein expression was significantly reduced in the FD group compared to the NC group (P = 0.008) and the SC group (P = 0.002; Figs. 1J, 1K). These findings indicated that the potassium channel gene Kir3.4 was involved in the development of FDM. 

KCNJ5 Overexpression Delays Myopia Progression

In order to ensure the stability of lentivirus transfection and the homology of experimental results, we extracted guinea pig RPE cells to construct KCNJ5 overexpressing lentivirus. The retinas were isolated from the eyes of 3-week-old guinea pigs and cultured with primary RPE cells, which were observed to become transparent as their pigment granules gradually disappeared (Fig. 2A). The cellular origin was determined by immunofluorescence staining, which indicated that RPE cells were positive for both intracytoplasmic keratin (pan-keratin) and S100B expression (Fig. 2B). 

Figure 2.

 

Guinea pig primary RPE cell culture and KCNJ5 overexpression lentivirus construction. (A) Changes in degranulation at each stage of 24 hours, 48 hours, 96 hours, and 120 hours of culture after isolation of primary RPE. Scale bar = 100 µm. (B) Results of immunofluorescence staining of RPE cells. Scale bar = 100 µm. (C) The expression of Kir3.4 at the mRNA level of cells in each group. (D) Western blot electrophoresis bands of Kir3.4 expressed in RPE cells. (E) The expression of Kir3.4 at the protein level of cells in each group. *P < 0.05, **P < 0.01, ***P < 0.001.

Following the transfection with KCNJ5-overexpression lentivirus, the expression of Kir3.4 at the mRNA and protein levels was examined in the RPE cells of each group. Compared with the NC and Vector groups, the expression of the KCNJ5-OE group was significantly higher at the mRNA level (all P_RPE < 0.001) and protein level (P_RPE = 0.004 and P_RPE = 0.012, respectively; Figs. 2C–E). The results showed that the lentivirus with KCNJ5 overexpression was successfully constructed. 

At the conclusion of transfecting the retina with lentivirus, the FD + KCNJ5-OE group exhibited lower refractive error (95% CI = 0.50–1.63) and shorter AL (95% CI = 8.14–8.22) compared with the FD + Vector group, the difference was statistically significant (all P < 0.001; Table 2; Figs. 3A, 3B). Hematoxylin and eosin showed that KCNJ5 overexpression inhibited retinal and choroidal thinning induced by FDM (P = 0.025 and P < 0.001, respectively; Figs. 3C–F). The efficiency of retinal transfection was validated by measuring the expression of Kir3.4 mRNA and protein, with results indicating a significant increase in the FD + KCNJ5-OE group compared with the FD + Vector group (P = 0.001 and P = 0.003, respectively; Figs. 3H–J). Given that the eye is a relatively closed organ, the likelihood of viral transfection into cells of other organs was considered to be very low, and no significant toxic side effects were observed in the animals during this experiment. 

Table 2.

 

Changes in SE and AL in Different Groups After Transfection of KCNJ5-OE LV (\(\overline{{\rm{X}}}\) ± s)

 

Figure 3.

 

Effect of KCNJ5 overexpression lentivirus transfected retina on FDM. (A) Changes in refractive state at 0, 1, 2, 3, and 4 weeks after lentiviral intervention. (B) Changes in AL at 0, 1, 2, 3, and 4 weeks after lentiviral intervention. (C) Results of hematoxylin and eosin staining of the retina in each group after 4 weeks. Scale bar = 50 µm. (D) Results of hematoxylin and eosin staining of the choroid in each group after 4 weeks. Scale bar = 50 µm. (E) Statistical results of total retinal thickness. (F) Statistical results of total choroidal thickness. (G) The levels of K⁺ concentration in the retinal tissue of each group after 4 weeks of lentiviral intervention. (H) Expression of Kir3.4 at the mRNA level in each group after 4 weeks of lentiviral intervention. (I) Western blot electrophoretic bands of Kir3.4 expressed after 4 weeks of lentiviral intervention. (J) Statistical results of stripe gray values in each group. *P < 0.05, **P < 0.01, ***P < 0.001.

The concentrations of K⁺ in the retinal tissues of the NC, FD, FD + Vector, and FD + KCNJ5-OE groups were examined. The results showed that the K⁺ concentrations in the FD + KCNJ5-OE group was lower than those in the FD and FD + Vector groups (all P < 0.001; see Fig. 3G). The changes in retinal K⁺ concentration and myopia have been consistent, suggesting that retinal K⁺ homeostasis plays an important role in myopia. 

M₂ Receptor Mediated Regulation of Potassium Concentration by Kir3.4

The M₂ receptor antagonist AF-DX 116 is a selective antagonist of the M₂ and M₃ subtypes of muscarinic receptors. It blocks the binding of acetylcholine to these receptors, thereby preventing the activation of downstream signaling pathways. It has been used in studies of the association of muscarinic receptors with myopia extensively and has been shown to inhibit the progression of myopia.^9,16,17 

LY2119620 is a selective agonist of M₂/M₄ receptors, with a modest 23.2% and 16.8% metabotropic agonism at M₂ and M₄ receptors, respectively. 

After the application of M₂ receptor inhibitors, the expression of CHRM2 at the mRNA and protein levels were determined within each group. The results indicated that compared with the NC group, CHRM2 expression was significantly reduced in the FD + AF-DX116 group (all P < 0.001), whereas it was significantly elevated in the FD + LY2119620 group (P < 0.001 and P = 0.021, respectively; Figs. 4A–C). These findings confirmed the effectiveness of the M₂ receptor intervention. 

Figure 4.

 

Effect of M₂ receptor interventions on CHRM2 receptor expression.(A) The expression of CHRM2 at the mRNA level in each group after 4 weeks. (B) Western blot electrophoretic bands of CHRM2 expressed after 4 weeks. (C) The expression of CHRM2 at protein level in each group after 4 weeks. *P < 0.05, ***P < 0.001.

Changes in the ocular bioparameters of the guinea pigs were observed at the end of the intervention. The development of myopia was found to be somewhat suppressed in the FD + AF-DX 116 group (SE, 95% CI = –1.19 to 0.99; and AL = 95% CI = 8.12–8.20) compared with the FD + DMSO and FD groups (all P < 0.001). Conversely, the FD + LY2119620 group showed a deeper degree of myopia with a significant increase in refractive error (95% CI = –5.41 to –4.29) and AL (95% CI = 8.43–8.57, all P < 0.001; Table 3; Figs. 5A, 5B). The retinal and choroidal thickness in the FD + AF-DX 116 group was thickened compared with the FD + DMSO group (all P < 0.001). The FD + LY2119620 group showed thinning of the retinal and choroidal thickness, which was at the same level as the retinal thickness in the FD + DMSO group (all P > 0.05; Figs. 5C–F). Additionally, we labeled apoptotic cells in the retina using TUNEL staining, whereas the presence of apoptotic cells was not detected in the retina of the intervention group (see Fig. 5H). The concentrations of the drug used in this study are all within the acceptable range for the animal. We did not observe adverse effects such as ocular inflammation in the pharmacological intervention group. 

Table 3.

 

Changes in SE and AL in Different Groups After Intervention M₂ Receptor (\(\overline{{\rm{X}}}\) ± s)

 

Figure 5.

 

The M₂ receptor intervention effects the development of myopia in vivo. (A) Changes in refractive state at 0, 1, 2, 3, and 4 weeks after intervention of the M₂ receptor. (B) Changes in AL at 0, 1, 2, 3, and 4 weeks after intervention of the M₂ receptor. (C) Results of hematoxylin and eosin staining of the retina in each group after 4 weeks. Scale bar = 50 µm. (D) Results of hematoxylin and eosin staining of the choroid in each group after 4 weeks. Scale bar = 50 µm. (E) The effect of drug injection on apoptosis of retinal cells in each group after 4 weeks. Scale bar = 50 µm. (F) Statistical results of total retinal thickness. (G) Statistical results of total choroidal thickness. (H) The levels of K⁺ concentration in retinal tissues of each group after 4 weeks of intervention with the M₂ receptor. (I) The expression of Kir3.4 at the mRNA level in each group after 4 weeks of intervention with the M₂ receptor. (J) Western blot electrophoretic bands of Kir3.4 expressed after 4 weeks of M₂ receptor intervention. (K) Statistical results of stripe gray values in each group. *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 6.

 

Schematic representation of the influence of potassium ion concentration in retinal on visual information transfer. (A) Light induction causes a transient increase in the concentration of K⁺ in the subretinal space, followed by an increase in the permeability of the Kir channels, contributing to the restoration of K⁺ in the subretinal space and maintaining the K⁺ homeostasis.⁴⁸ The stable K⁺ concentration can maintain the precise transduction of photoreceptors to light stimuli. (B) The RIDE model proposes that masking leads to the accumulation of K⁺ in the subretinal space. In the present study, it was observed that the reduced expression of Kir3.4 following form deprivation led to the re-accumulation of K⁺ and the dysregulation of K⁺ homeostasis, which subsequently affected photoreceptor excitability and ultimately resulted in refractive error (figure created with BioRender).

The relative expression of Kir3.4 mRNA was significantly upregulated in the FD + AF-DX 116 group compared to the FD + DMSO group (P < 0.001), which was also observed in the FD + LY2119620 group (P = 0.002), although this increase was substantially lower than that observed in the FD + AF-DX 116 group (P < 0.001; Fig. 5I). Further examination of the expression of the Kir3.4 protein revealed that there was a significant upregulation in the FD + AF-DX 116 group (P < 0.001) compared with the FD + DMSO group. The FD + LY2119620 group displayed that the difference was not statistically significant compared with the FD + DMSO group (P = 0.092; see Figs. 5J, 5K). 

The K⁺ concentrations within each group were examined using a potassium ion assay kit, and it was found that the K⁺ concentrations were significantly lower in the FD + AF-DX 116 than in the FD and FD + DMSO groups (P = 0.008 and P = 0.004, respectively). In contrast, the K⁺ concentrations were significantly increased and much higher levels in the FD + LY2119620 group than those in the FD and FD + DMSO groups (all P < 0.001; see Fig. 5G) The changes of K⁺ concentration and Kir3.4 expression in the FD + AF-DX 116 group were consistent with the change of KCNJ5 in the overexpression group. Similar refractive changes were also observed in the two groups, suggesting that kir3.4 was the target of M₂ receptors. 

Discussion

This is the first in vivo study implicating the association between muscarinic receptor M₂ and the potassium channel Kir3.4 in the development of myopia. M₂ receptor antagonist AF-DX 116 retarded myopia progression in experimental myopic guinea pigs, and M₂ receptor intervention was also found to have a regulatory effect on the K⁺ transporter, which was determined to be critical in influencing the development of myopia. In the present study, we found that selective M₂ receptor antagonists significantly delayed myopia progression by modulating Kir3.4 channels and retinal K⁺ homeostasis. Compared with atropine, such drugs can avoid pupil dilation and regulatory paralysis, providing a more precise intervention strategy for myopia prevention and control in children. M₂R/Kir3.4 dual-targeting drugs may be developed in the future to further improve efficacy and safety. 

Role of MRs in Myopia

The concept of acetylcholine (ACh) was first introduced by Henry Dale in 1914, when he demonstrated its essential role in neuronal transmission. ACh functions as a neurotransmitter that transmits neuronal signals by binding to cholinergic receptors, with MRs being the primary receptors for ACh. In the late 1980s, researchers successfully identified 5 distinct subtypes of MRs using molecular biology techniques. Goyal¹⁸ systematically discussed the structure, function, and binding properties of muscarinic receptors M₁ to M₅. The use of atropine for the treatment of myopia was first reported by Wells in the 19th century, and a steady stream of studies has begun to focus on the potential effects of atropine on myopia.¹⁹ Bedrossian reported that atropine effectively prevented the progression of myopia in eyes treated with long-term eye drops. Kennedy used a 1% atropine solution to treat 214 patients with myopia and found that it slowed the progression of myopia after 3.5 years of follow-up. However, early studies noted side effects such as accommodation paralysis, pupil dilation, and drug toxicity due to prolonged atropine use, which hindered its clinical application. The National Eye Center in Singapore compared the efficacy and adverse effects of different concentrations of atropine for myopia control in a large clinical trial, revealing that 0.01% atropine offered the optimal balance of efficacy and safety for this purpose.²⁰ 

Among the numerous strategies available for the prevention and control of myopia, atropine has emerged as an effective intervention, particularly for children. Several randomized controlled trials have demonstrated the safety and efficacy of topical low concentrations of atropine for myopia control.^3,4,21 Barathi et al.⁹ suggested that myopia progression may be mediated by the M₂ receptor and proposed that M₂ receptor subtypes could influence myopia progression through ion channels or be activated by second messengers or direct receptor action. The present study builds on this foundation and examines in detail the relationship between the M₂ receptor and the potassium channel Kir3.4 in the development of myopia. 

Role of Kir3.4 in Myopia

The function of ion channels in the retina is critical for maintaining the intracellular electrochemical reactions, mechanical force transduction, and visual cycle in both the light and dark components of the central nervous system as well as the neuro-visual components.^22–24 The difference between the resting membrane potential and the potassium balance potential of RPE cells provides the driving force for K⁺ flux and sustains a stable concentration of K⁺ in the subretinal lumen, which is essential for maintaining photoreceptor excitability.²⁵ The retinal ion-driven fluid efflux (RIDE) model proposes that FDM is due to altered visually driven phototransduction due to masking, which reduces ion-driven fluid transport from the vitreous across the RPE to the choroid.^26–30 The RIDE model further proposes that masking decreases light stimuli entering the eye, leading to the reaccumulation of K⁺ in the subretinal lumen.^31–34 Biochemical and proteomic analyses have revealed an altered electrolyte balance and a significant decrease in vitreous fluid K⁺ concentration among highly myopic populations.³⁵ Skarphedinsdottir et al.³⁶ demonstrated that light-induced increases in subretinal K⁺ transient transport are dependent on potassium ion channels. Our team previously demonstrated that the concentration of K⁺ in the retina of FDM eyes was significantly higher than that observed in NC eyes of guinea pigs.³⁷ 

The results of the study suggest that the overexpression of KCNJ5 reduces retinal K⁺ concentration and slows the rate of AL growth, thereby delaying the progression of myopia. The accumulation of K⁺ in an extracellular environment may play a physiologically important role in reducing muscle excitability,³⁸ and is implicated in both the development of muscle fatigue and blood flow regulation.^39,40 Previous studies have shown that dysregulation of K⁺ homeostasis within the ciliary muscle leads to poor ciliary muscle contraction during visual accommodation in guinea pigs, resulting in underaccommodation that promotes the onset and progression of myopia.¹⁴ Inwardly rectifying potassium channels help epithelial cells in transporting K⁺ from the subretinal lumen across the RPE into the choroidal circulation, thereby maintaining normal K⁺ homeostasis.²³ We speculate that this is the reason why overexpression of KCNJ5 can delay the development of myopia. These studies suggest that changes in K⁺ homeostasis may be involved in ocular growth and the formation of myopia. Kir3.4 may be involved in the maintenance of photoreceptor excitability as well as in critical processes, such as retinal K⁺ transport that are critical for structural changes in myopia. The downregulation of Kir3.4 in the retinal tissue of myopic eyes suggests that it may serve as a potential therapeutic target for controlling myopia progression (Fig. 6). 

Regulation of Kir3.4 by M₂ Receptor

Kir3.4 is a member of the GIRK channel family, but it does not promote the influx of K⁺, as its name might imply. GIRK permeability increases when the membrane potential transitions from depolarization to hyperpolarization, K⁺ efflux is inhibited or may be completely halted, resulting in inward rectification. GIRK is crucial for maintaining the resting potential and regulating neuronal excitability by coupling with various receptors (e.g. M₂ muscarinic receptor, GABAB receptor, opioid receptors, and D₂ dopamine receptor).^12,41–43 Kir3.4 channels can be activated by G proteins, PIP₂, intracellular sodium ion, ethanol, and mechanical stretch,⁴⁴ but classical G protein-coupled signaling still predominates in organismal tissues. Furthermore, the M₂ receptor has been demonstrated to preferentially couple with the Pertussis toxin (PTx)-sensitive G proteins G_i and G_o^45,46 (Fig. 7). 

Figure 7.

 

Schematic representation of M₂ receptor signal transduction and GIRK4 channel activation. (A) Agonist binding promotes the formation of a GPCRs -Gα (GDP) βγ complex. The activated GPCRs then triggers the exchange of GDP to GTP on the Ga subunit. The Gα (GTP) and Gβγ subunits subsequently dissociate from the GPCRs. Dissociated Gβγ directly binds to and activates GIRK channels. Dissociated Gα (GTP) hydrolyzes GTP to GDP, which then reassociates with Gβγ to form Gα (GDP) βγ. (B) The M₂-receptor belongs to the superfamily of GPCRs, but has an inhibitory effect of M₂-receptor binding to Gαi/o. When the antagonist (AF-DX 116) binds to the M₂ receptor, the GIRK4 channel is turned on, and the agonist (LY2119620) function in the opposite direction (figure created with BioRender).

In the M₂ receptor intervention model, the Kir3.4 protein was upregulated, whereas the retinal K⁺ concentration was decreased in the FD + AF-DX 116 group. Conversely, the opposite effect was observed in the FD + LY2119620 group. The permeability of the Kir3.4 channel decreased with membrane potential depolarization and increased with hyperpolarization. AF-DX 116 increased the permeability of the Kir3.4 channel, preventing excessive K⁺ efflux in the subretinal space and ultimately slowing down the progression of myopia. These results indicate that muscarinic antagonists inhibit the progression of myopia by altering the permeability of potassium channels, thereby regulating K⁺ transport within the eye. 

However, given that the choroidal thickness of the FD + AF-DX 116 group was significantly higher than that of the FD group (see Figs. 5D, 5F), and M₂ expression was also present in the choroid, we speculated that AF-DX 116 may also play a role in delaying myopia by relaxing choroidal blood vessels to increase blood flow, which can be explored in the future through more detailed experimental designs. 

Disputes and Limitations

The muscarinic receptor antagonist functions by reversibly blocking muscarinic receptors on postganglionic cholinergic innervated effectors and are currently prominent in both domestic and international research aimed at intervening in the development of myopia, as well as one of the best therapeutic modalities to delay the development of myopia at present. However, the specific site of action of MRs is still controversial. Some researchers believe that MRs target the sclera and affect scleral tissue proliferation and matrix synthesis, thereby affecting scleral remodeling. Conversely, others propose that MRs target the retina and affect retinal signaling by modulating the release of various neurotransmitters from organelles, thereby controlling ocular development and inhibiting the development of myopia. McBrien et al.⁴⁷ analyzed and found that revealed varying concentrations of atropine low, medium, and high demonstrated efficacy in inhibiting myopia development in both animal and human studies, but the low concentrations of atropine did not reach the levels necessary to induce changes in scleral tissue. Additionally, Mitchelson et al.⁷ utilized quantitative RT-PCR to determine that the M₃ receptor mRNA accounted for 76% of human scleral tissue, and M₁, M₂, M₄, and M₅ receptor mRNA accounted for 23%, 0.007%, 0.060%, and 0.700%, respectively. This study indicated that interference with the M₂ receptor to regulate retinal potassium channels effectively delayed the development of myopia. Therefore, the present study speculates that the muscarinic receptor may influence scleral remodeling by affecting retinal signaling and may also regulate ciliary muscle function by modulating ion homeostasis. 

In this study, 3-week-old male guinea pigs were selected as the myopia model animals. As mammals, the refractive development of guinea pigs closely resembles that of humans. The prevention and control of myopia primarily target school-age children, and 3-week-old guinea pigs are within the critical period of visual development, which can simulate the developmental changes of myopia in children and enhance the feasibility of translating basic experimental findings to clinical applications in the future.⁴³ However, guinea pigs lack the fovea of primates and have a smaller vitreous to retinal ratio, which may result in different K⁺ dynamics than in humans. In addition, as diurnal mammals, guinea pigs may have a retina that is more sensitive to form deprivation than humans. Barathi et al.⁹ reported a more effective myopia inhibition using 0.1% AF-DX 116 eye drops in mice. The comparatively lower degree of myopia inhibition observed with AF-DX 116 in the present study may be due to differences in concentration, target subjects, and the route of administration. LY2119620 exhibits modest 23.2% and 16.8% metabotropic agonism of the M₂ and M₄ receptors, respectively. This may explain the lack of significant inhibition in Kir3.4 protein expression within the FD + LY2119620 group. Furthermore, several limitations were identified in this study. Multiple drug concentration gradients were not established during the experiment, and subsequent changes in K⁺ concentration within the vitreous cavity and choroid were not detected. We cannot rule out the roles of other tissues, further studies will be necessary to investigate the effects of muscarinic receptors in different tissues on myopia development. 

Conclusions

In summary, we investigated the mechanism by which muscarinic M₂ receptor antagonists delay the progression of myopia. The M₂ receptor antagonists delay myopia progression by regulating retinal K⁺ via the G protein/Kir3.4 pathway. Overexpressing KCNJ5 reduced K⁺ concentration and attenuated myopia, suggesting a key role for retinal K⁺ homeostasis in myopia. The above results reveal a novel mechanism for M₂ receptor targeted interventions and provide an important basis for the clinical search for target drugs for myopia prevention and control. 

Acknowledgments

Supported by Natural Science Foundation Project of Science & Technology Department of Sichuan Province (No. 2023NSFSC0595), and Key Project of the Affiliated Hospital of North Sichuan Medical College (No. 2023ZD010). 

Author Contributions: H.Z., Performed the experiments, analyzed the data, and wrote the manuscript; G.Z. and Q.Y., Interpreted the results; J.N., R.W., H.L., and S.H., Provided technical and material support; H.B. and X.L., Designed the study and reviewed and revised the paper. 

Data Availability Statements: Related research datasets of this study are available from the first authors upon reasonable request. 

Disclosure: H. Zhou, None; G. Zhou, None; Q. Yang, None; J. Niu, None; R. Wang, None; H. Liu, None; S. Hou, None; H. Bi, None; X. Lia, None