All animal experiments in this study were approved by the Animal Experimental Committee of Keio University (permit number: 16017). Our study adhered to the Institutional Guidelines on Animal Experimentation at Keio University, the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research, and Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines for the use of animals in research. We purchased male C57BL/6 J mice (three weeks old) from CLEA Japan (Yokohama, Japan). Five mice were maintained in one cage with free intake of standard chow and water. We kept them in an environment with a 12 h/12 h light/dark cycle (the dark cycle extended from 8:00 p.m. to 8:00 a.m.) at 23 ± 3 °C.

According to the manufacturer’s instructions, Tm (Cayman Chemical, Catalogue #:11445) was dissolved in dimethyl sulfoxide (DMSO) at a concentration of 100 mg/mL, stored at − 20 °C, and diluted 1000-fold with phosphate-buffered saline (PBS) immediately before use to prepare eyedrop solutions of 25, 50, and 100 μg/mL. Mice were treated with different concentrations and/or a different number of applications of Tm eyedrops, and PBS containing the same concentration of DMSO was used as the control, 5 μL per eye. Ocular components were measured, and scleral tissues were collected for the assays on days 0, 7 and 21 after treatment. The experimental plan for the animals is shown in Additional file 1: Fig. S1.

AL was measured from the anterior corneal surface to the retinal pigment epithelium using a spectral domain optical coherence tomography (SD-OCT) system (Envisu R4310, Leica, Germany). The refraction was detected using an eccentric infrared photo refractor (Steinbeis Transfer Center, Germany) at the vertical pupil meridian. The data were automatically recorded by the program when the parameter values were stable. Choroidal thickness (ChT) was measured using SD-OCT, and the posterior surface of the choroid was quantified using Image J (Ver 1.53, NIH). ChT was determined using the formula: area divided by circumference. Intraocular pressure (IOP) was measured using a tonometer (Tono Lab, Icare Finland Oy, Vantaa, Finland) calibrated for mice under anaesthesia. Parts images of ocular components measurement were shown in Additional file 5: Fig. S5.

Mice were anesthetized with 0.75 mg/kg medetomidine (Sandoz K.K., Tokyo, Japan), 4 mg/kg midazolam (Domitor®, Orion Corporation, Espoo, Finland), and 5 mg/kg butorphanol tartrate (Meiji Seika Pharma Co., Ltd., Tokyo, Japan) dissolved in normal saline. Mydriasis was induced by 0.5% tropicamide eye drops (Santen, Osaka, Japan), and mice were placed in a cylindrical holder for measurement.

The eyeballs were harvested after euthanising the mice and fixed in Super Fix (KURABO, Osaka, Japan) for three days at 4 °C with gentle shaking. The fixative solution was changed every day. After fixation, paraffin sections were prepared as previously described [6]. To measure collagen content, paraffin sections were stained using a picrosirius red stain kit (Polysciences, Inc., USA) according to the manufacturer’s instructions. In brief, sections were de-paraffinized and hydrated with deionised water (DI) and then placed into different solutions in the following order: solution A (phosphomolybdic acid) for 2 min, followed by rinsing with DI water; solution B for at least 60 min; and solution C for 2 min. Finally, the sections were dehydrated and cleared using 70% ethanol for 45 s. The sections were visualised under bright field and polarised light (Olympus BX53). Collagen shows birefringence under polarised light, and picrosirius red staining can demonstrate both bundled and unbundled collagen fibres. All reconstructed images were converted to 8-bit using ImageJ (v.1.53, NIH). A threshold was set to distinguish and analyse collagen size and the size of collagen fibres. Collagen fibres with sizes above the threshold were considered bundled collagen fibres. It is worth noting that the threshold value remains the same for all images analysed. Pixel intensity was measured thereafter (Additional file 1: Fig. S4).

TUNEL assay was performed using an in situ Apoptosis Detection Kit following the manufacturer’s protocol (TAKARA Bio, Shiga, Japan) and images were captured with a BZ-X800 system (Keyence, Tokyo, Japan).

After experimental intervention and measurement of ocular parameters, eyeballs were enucleated from C57BL/6 J mice. For protein expression analysis, isolated cornea, lens, retina, choroid and sclerae from eyeballs were homogenized in RIPA buffer (50 mM HEPES (pH 7.5), 150 mM NaCl, 1% NP-40, 0.1% sodium deoxycholate, 1 mM EDTA, 5 mM benzamidine, 10 mM β-glycerophosphate, 1 mM Na3VO4, 50 mM NaF, and 1 mM PMSF) containing Halt protease inhibitor cocktail (ThermoFisher Scientific, Waltham, MA, USA). After centrifugation, protein concentration was measured using the BCA (Pierce BCA Protein Assay Kit, Thermoscientific, Waltham, MA, USA) method and adjusted to 1.0 g/L with Laemmli sample buffer (Nacalai Tesque, Kyoto, Japan). Samples were stored at − 30 °C until further use.

To visualize protein expression, sodium dodecyl sulphate–polyacrylamide gel electrophoresis was performed using 10% acrylamide gels with protein-size markers (MagicMark XP Western Protein Standard, ThermoFisher Scientific). Proteins were transferred to polyvinylidene fluoride membranes (Merck Millipore, MA, USA), blocked with Blocking One (Nacalai Tesque, Tokyo, Japan), and incubated overnight at 4 °C with the following antibodies: anti-ATF6 (24169-1-AP, Proteintech, CHI, USA), IRE1 alpha (phosphor Ser724) (GTX132808, GeneTex, CA, USA), anti-col1-3/4 mAb (0217-050, Immunoglobe, Germany), COL1A1 Rabbit mAb (#84336), IRE1 (14C10) Rabbit mAb (#3294), phosphor-PERK (Thr980, 16F8) Rabbit mAb (#3179), PERK (C33E10) Rabbit mAb (#3192), MMP2 Rabbit mAb (#4022), Bip (C50B12) Rabbit mAb (#3177), phosphor-eIF2α (Ser51) (D9G8) XP Rabbit mAb (#3398), eIF2α (D7D3) XP Rabbit mAb (#5324),and β-actin (8H10D10) Mouse mAb (#3700) (Cell Signaling Technologies Japan, Tokyo, Japan). Membranes were incubated with the appropriate horseradish peroxidase-conjugated secondary antibodies and visualised using SuperSignal West Femto Maximum Substrate (ThermoFisher Scientific, Waltham, MA, USA). After visualisation, the membranes were incubated with Restore PLUS Western Blot Stripping Buffer (ThermoFisher Scientific, Waltham, MA, USA) and re-incubated with another primary antibody. Results for phosphorylated and total, and target and β-actin antibody stripping used the same membrane. Densitometric analyses were performed using ImageJ software (v.1.53, NIH). Uncropped blots in this manuscript were shown in Additional file 3: Fig. S3.

Differences between the control and experimental groups were compared using Student’s t-test or One-way analysis of variance with least significant difference (LSD) post hoc test to calculate statistical significance (GraphPad Prism software, v.8.0). All data are expressed as mean ± SD. Statistical significance was set at P < 0.05.