In Vitro Effect of the Biomechanics After Photodynamic Therapy (PDT) Intervention on Rabbit Cornea

Objective: To observe the change of the biomechanical properties of rabbit cornea after the intervention by corneal collagen cross linking (CXL) and toluidine blue O combined with red light (TBOR). Methods: The study was carried out in compliance with the ARRIVE guidelines. 20 healthy adult New Zealand white rabbits were randomly divided into two groups. One group was taken with Riboavin combined with UV light therapy in the right eye of healthy New Zealand white rabbits, and the other group was taken toluidine blue O combined with red light (TBOR) treatment. All left eyes were taken as controls. Parameters, from Pentacam and Corvis ST like K1 (The keratometry readings of the attest), K2 (The keratometry readings of the steepest), ACD (Anterior chamber depth), Pupil diameter, ACA (Mean angle of the anterior chamber), ACV (Volume of the anterior chamber), bIOP (Biomechanical Intraocular pressure), CT (Corneal thickness, pachymetry), A1 Time(Time from starting until the rst applanation), A1 Velocity(Corneal speed during the rst applanation moment), A2 Time (Time from starting until the second applanation), A2 Velocity (Corneal speed during the second applanation moment), HC Time(Time from starting until highest concavity is reached), Peak Dist(Distance of the two “peaks” at highest concavity) and HC-Radius(The radius at highest concavity), were examined before and 2 weeks after intervention. The rabbits were sacriced after anesthesia. Then, their corneas were removed for corneal stretch test. Results: With the examination of Pentacam and Corvis ST, IOP(11.28 ± 11.2mmHg v.s. 6.66 ± 4.02mmHg) and A1 time(7.03 ± 1.27s v.s. 6.55 ± 0.35s) were increased, comparing with those before intervention. From the in vitro corneal stretch test, the tangent modulus of the CXL group was more than 3 times of the Control group, whereas the tangent modulus of the TBOR group was about 0.7 of the biomechanical indexes of rabbit corneas by two kinds of PDT CXL and TBOR) before and after the intervention. Results show that after the intervention of the two methods, Et value of CXL group increased signicantly compared to the TBOR group and the control group. Using animal model, we veried the clinical effectiveness of using CXL in slowing the progressive of keratoconus 21, 22, 32 . However, TBOR therapy did not increase the biomechanical properties of the cornea. Future investigators should be aware that TBOR may not affect biomechanical properties of the cornea.


Introduction
With unique structure 1, 2 and biomechanical properties, 3 4 cornea can maintain the transparency and provide a clear vision. In recent years, corneal collagen cross linking (CXL) was applied in the eld of the treatment of keratoconus and infectious corneal disease. This technique can not only kill bacteria but also can increase corneal rigidity and the resistance of degradation ability of enzyme. 5 In vitro experiments conducted by Martins et al. 6 demonstrated corneas intervened by ribo avin combined with UVA(CXL) could inhibit the growth of bacterial drug-resistance. Tissues targeted by irradiation of special light causes transformation of oxygen into ROS (reactive oxygen species) in the aerobic environment around the excited state energy, this mechanism can be applied many different organelles producing toxic effect which causes cell death. 7 Different light sources, such as Ultraviolet( 400 nm), 8 blue light(440 ~ 480 nm), 9 green light(500 ~ 560 nm) 10 and red light(600 ~ 780 nm) 11 combined with different photosensitizers such as ribo avin, 12 Rose Bengal, 10 5-aminolevulinic acid, 13 Chlorin e6 14 and toluidine blue 11 can create different types of PDT. In previous study, our group had con rmed that CXL 15 and TBOR 16 had an apparent inhibitory effect on bacterial growth through in vitro studies. In this study, we will investigate the effects of PDTs on biomechanics properties of corneas by biomechanical examination following PDT treatments.

Experimental animals
The study was carried out in compliance with the ARRIVE guidelines. All operations involving animals were carried out in accordance with the standard established by the ARVO Statement for Use of Animals in Ophthalmic and Vision Research. Our study protocol was approved by the Animal Care and Use Committee, Capital Medical University (AEEI-2017-081). Twenty healthy adult New Zealand white rabbits (both genders, aged 3-4 months, weight 2.5-3.0 kg) were provided by the Beijing Longan Research Animal Center (License NO. SCXK Beijing 2-14-0003). The right eye of each animal was selected as the experiment group, whereas, left eye was taken as the a control group.

Experimental design
Twenty New Zealand white rabbits were randomly divided into two groups. Ten in each of the CXL and TBOR groups. Ribo avin (0.1wt%) combined with UV therapy (5.273mw /cm2,30min) and toluidine blue O (1mmol/L) combined with red light (87.8mw /cm 2 ,30min) were applied to intervene in the right eye grouping of New Zealand white rabbits individually ( Figure 1).

In vivo biomechanical examination
The corneas of each group were examined by Pentacam (Oculus, Wetzlar, Germany) and Corvis ST (Oculus, Wetzlar, Germany) at different time points (before intervention and 2 weeks after intervention).
During the examination, two assistants were required. One assistant wrapped the surgical towel around the experimental rabbit with only the head exposed. The experimental eyes were placed in front of the test lens following the adjustment of its position and direction until a clear visual appears on the monitor.
Interventions will then be carried out. Detailed parameters are shown in Table 1 and Table 2.
In vitro biomechanical testing Two weeks after the interventions, rabbits were euthanized with an intravenous overdose injection of 5% pentobarbital (1 mg/kg). Both eyes of each rabbit were immediately enucleated. The corresponding 20 left eyes constituted the control group. Cornea is removed using an ophthalmic scalpel and interstitial scissors along the periphery of the rabbit cornea about 2-3mm away from the sclera at room temperature.
Other ocular components were also removed. Corneal strips of about 3mm width were made with a two-parallel surgical blades 19 (Figure 2A) in the inferior-superior direction. (Note that the corneal strips should have 2-3mm sclera at both ends) The strips were connected to a pair of mechanical clamps, leaving 10mm between the two clamps. Then material testing machine (EZ Test Compact Table-Top Universal Tester, Shimadzu, Kyoto, Japan) was used for uniaxial tension test on the corneas ( Figure 2B). The sclera part at both ends of the cornea strip was clamped on both ends. The machine was equipped with a 50 N capacity load cell, the test was conducted at room temperature. The humidity around the test was reduced by air humidi er to simulate the tissue environment. The initial distance between the two clamps with a Vernier caliper was measured and recorded it as Lo. We carried out four loading and unloading cycles on samples with the 1mm /min tensile rate and 0.10N as the maximum load ( Figure 2C). The behavior recorded in the fourth cycle was considered to be representative of the sample stability results. 20 The load-displacement (F -△L) data for the fourth cycle was used to calculate the stress under each load(F) , stress is σ=F/wt (t was the average corneal thickness and w is the average specimen width). The related strain was obtained as = △L/Lo. The result of stress-strain conforms to the tting functionexponential function σ =A(e B -1), where A and B are constants. The tangent modulus Et is calculated as 19 .

Statistical analysis
All analysis was performed using the SPSS 24.0 (SPSS Inc., Chicago, USA). Comparisons between biomechanical parameters in the two different groups were performed using the paired T-test. P values 0.05 were considered statistically signi cant.

Uniaxial tension test
As shown in Figure 3, the curves of load-displacement diagrams of the three groups are obviously different, which was determined by parameters A and B (Table 3). Through the stress-strain (σ-) relationships (Figure 4), the tangent modulus (Et= dσ/d ) can be obtained at any stress level. The tangent modulus (Et values) of the three groups (10, 20 and 30kPa stresses) were varied signi cantly (

Discussion
Our study attempted to address the biomechanics of cornea after different PDT treatments. Results demonstrated that CXL intervention resulted in signi cantly increase in the mechanical stiffness (as measured by tangent modulus, Et) of the ex-vivo rabbits' cornea. However, the effect of TBOR method on corneal hardness was not established. The changes in corneal biomechanics in vivo were not signi cant.
Little attention has been given to the effect of PDT on cornea biomechanics. Previous studies on the biomechanical properties of cornea were largely CXL clinical studies using Pentacam, [21][22][23] Corvis ST 23,24 and ORA(Ocular response analyzer) 25,26 . Nevertheless, the biomechanics metrics provided by Pentacam and Corvis ST cannot be linked directly to the traditional measures of tissue stiffness (here is mainly Et). At present, most of the PDT studies on cornea focus on CXL technique. It showed that CXL can increase the corneal hardness and prevent the development of keratoconus. 21,27 CXL has also been shown to increase the corneal hardness in a number of animal studies. 28, 29 In addition, current studies only used traditional Young's modulus to study changes in the biomechanical properties of the cornea in vivo. 27 Tangent modulus (Et) refers to the ratio of instantaneous stress and instantaneous strain of a tissue material at a some timepoint. 30,31 Et value is rarely used as to evaluate the change of biomechanical property in cornea. While there is no agreement on the effect of cornea's rigidness after CXL. With Et values analysis method, we found that CXL could effectively increase the hardness of rabbit cornea. This conclusion was also consistent with the study of Wollensak et al. 28 and others. 27,29,31,32 . Our previous studies demonstrated the effectiveness of TBOR on the treatment of bacterial keratitis. 11,16 However, few researchers have studied the effect of TBOR on corneal biomechanics, or comparing the Et value of CXL and TBOR on corneal biomechanics.
In this study, corneal hardness was signi cantly increased after CXL intervention, while TBOR had little effect on corneal hardness. TBOR's Et values were lower to that of the control group. According to the optical knowledge, red light wavelength is about 630 nm, UV wavelength is about 370 nm, c = λ* f (c is the speed of light, λis wavelength, f is frequency), the longer the wavelength, the lower the frequency.
According to the formula "E = h * f (E is energy, h is Planck's constant)", the higher the frequency, the higher energy and the stronger penetrating power. Therefore, the higher UV energy has stronger penetrating power. Based on these facts, we believe that UV light can easily penetrate epithelial layer of the cornea, reach the stroma layer and induce the matrix layer ber to arrange more closely, thus harden the cornea.
Our studies demonstrated the change of the biomechanical indexes of rabbit corneas by two kinds of PDT methods (both CXL and TBOR) before and after the intervention. Results show that after the intervention of the two methods, Et value of CXL group increased signi cantly compared to the TBOR group and the control group. Using animal model, we veri ed the clinical effectiveness of using CXL in slowing the progressive of keratoconus 21,22,32 . However, TBOR therapy did not increase the biomechanical properties of the cornea. Future investigators should be aware that TBOR may not affect the biomechanical properties of the cornea.
This study has some limitations. Firstly, our sample size was limited to only 10 rabbits in each group. So larger sample size maybe necessary to further study the effect of CXL and TBOR or cornea biomechanics. Secondly, our tests were done ex vivo, we tried to preserve the cornea tissue and examine them within 2 hours post-mortem, degradations of the cornea might still occur, which might lead to inaccurate results. Thus, further tests should be done with the limitations in mind.
Overall, we found that CXL can effectively increase the biomechanical properties of cornea and

Consent for publication This study was published with the consent of all authors
Availability of data and materials The datasets used during the current study are available from the corresponding author and rst author on reasonable request.
Authors' Contributions: Qingfeng Liang directed the whole process of the study and sponsored the study; Guanyu Su was the main author of the study and wrote the paper; Shaofeng Han was the author of Figure 3, Figure 4, Table 3 and Table 4 of the study; Bowen Song polished the language of the paper; and Kai Cao was the statistical analysis of the study.

Parameters
Means unit

K1
The keratometry readings of the attest D

K2
The keratometry readings of the steepest D     Mean load-displacement behavior in three groups.

Figure 4
Mean corneal stress-strain behavior of corneas in three groups, error bars represent the standard deviation of strain values.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.