You are on page 1of 6

Experimental Eye Research 165 (2017) 1e6

Contents lists available at ScienceDirect

Experimental Eye Research


journal homepage: www.elsevier.com/locate/yexer

Riboflavin and ultraviolet A irradiation for the prevention of


progressive myopia in a guinea pig model
Xiaoxia Li, Miaoqin Wu*, Luyi Zhang, Hui Liu, Lan Zhang, Jinjing He
Department of Ophthalmology, Zhejiang Provincial People's Hospital, People's Hospital of Hangzhou Medical College, Hangzhou, Zhejiang, China

a r t i c l e i n f o a b s t r a c t

Article history: In this study, we evaluated the effect of oral administration of riboflavin combined with whole-body
Received 26 September 2016 ultraviolet A (UVA) irradiation on the biochemical and biomechanical properties of sclera in a guinea
Received in revised form pig model to control the progression of myopia. Experimental groups were administered 0.1% riboflavin
27 August 2017
solution with or without vitamin C by gavage from 3 days before myopic modeling and during the
Accepted in revised form 28 August 2017
modeling process. Guinea pigs underwent 30 min of whole-body UVA irradiation after each gavage for 2
Available online 31 August 2017
weeks. For control groups, guinea pigs were administered vitamin C and underwent either whole-body
UVA irradiation without 0.1% riboflavin solution or whole-body fluorescent lamp irradiation with or
Keywords:
Myopia
without 0.1% riboflavin solution. Resultantly, myopia models were established with an increased axial
Riboflavin length and myopic diopter. Compared with myopic eyes in the control groups, the net increase in axial
Sclera length, diopter and strain assessment decreased significantly, and the net decrease in sclera thickness,
UVA ultimate load, and stress assessment decreased significantly in experimental groups. MMP-2 expression
showed a lower net increase, while TIMP-2 expression showed a lower net decrease. In addition, hy-
perplasia of scleral fibroblasts was more active in myopic eyes of experimental groups. Overall, our re-
sults showed that oral administration of riboflavin with whole-body UVA irradiation could increase the
strength and stiffness of sclera by altering the biochemical and biomechanical properties, and decreases
in axial elongation and myopic diopter are greater in the guinea pig myopic model.
© 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND
license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction that the sclera behaves as a “metabolic disorder” (McBrien and


Gentle, 2003). Various inborn and postnatal factors cause the
Myopia is a significant public health problem with an increasing sclera to be thin and expand under eye pressure, leading to eyeball
prevalence, and it is the leading cause of correctable visual deformations such as localized ectasia of the posterior sclera, which
impairment in children and adults in the developed world and one results in myopia. Animal studies of experimental myopia have
of the main causes of blindness in developing countries (Robaei shown that scleral tissue from myopic eyes decreases mechanical
et al., 2006; Lu et al., 2009; Vitale et al., 2008; Resnikoff et al., stiffness and increases creep compliance compared with scleral
2008). In China, there is a high prevalence of myopia, with an tissue from normal eyes (Phillips et al., 2000). The concept of
average morbidity of more than 75% in young adults aged 15e25 enhancing the strength and stiffness of the sclera to stabilize the
years (Holden et al., 2014). normal morphology of ocular tissue has attracted significant
Myopia, particularly high myopia, is associated with major research interest in recent years.
ophthalmic diseases that can eventually lead to blindness, such as Wollensak and Iomdina reported stressestrain measurements
exudative myopic macular degeneration, rhegmatogenous retinal of a sector of the equatorial sclera of Chinchilla rabbit eyes, which
detachment, myopic retinopathy, and glaucoma. The mechanism of was treated in vivo with ultraviolet A (UVA) irradiation and ribo-
myopia is not completely understood, but many researchers believe flavin drops, and proposed that scleral collagen cross-linking
induced by UVA and riboflavin may increase biomechanical
strength and scleral rigidity (Wollensak and Iomdina, 2009).
* Corresponding author. Department of Ophthalmology, Zhejiang Provincial
Another study investigated the in vitro ultrastructural effects of a
People's Hospital, People's Hospital of Hangzhou Medical College, 158 Shangtang collagen cross-linking treatment with riboflavin instilled into the
Road, Hangzhou, Zhejiang 310014, China. strips prior to UVA irradiation in human corneo-scleral collagen
E-mail address: eyewmq@126.com (M. Wu).

http://dx.doi.org/10.1016/j.exer.2017.08.019
0014-4835/© 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
2 X. Li et al. / Experimental Eye Research 165 (2017) 1e6

fibrils and found that this method causes structural changes in the establishing myopia models to two weeks after the modeling.
collagen fibril network of the sclera (Choi et al., 2013). Riboflavin- During the modeling period, animals underwent fluorescent lamp
sensitized photoreaction generates free radicals and reactive oxy- irradiation (30 cm from above) or UVA irradiation (370 ± 5 nm)
gen species, such as singlet oxygen, superoxide, or superoxide using a UVA device (Changjiang Medical Devices Co., Ltd, China)
anion radicals to induce cross-linking of collagen under UVA with an irradiance of 3.0 mW/cm2 at a distance of 30 cm for whole-
(Wollensak and Spoerl, 2004; Wollensak et al., 2005). Scleral body irradiation. For fluorescent lamp irradiation, the irradiation
collagen cross-linking by riboflavin/UVA is a novel technique to started after each gavage and the total daily irradiation time was
enhance scleral mechanical strength, and may provide an ideal 8 h; for UVA irradiation, the irradiation started after each gavage
approach for strengthening scleral tissue, thereby suppressing and lasted for 30 min, and the total daily irradiation time was 1.5 h.
myopic progression. However, these experiments using riboflavin The right eye of each animal served as the lens-induced eye and the
and UVA require a surgical operation to expose the sclera with a contralateral left eye served as the control eye.
limited irradiation area. At this time, further studies are required
for this intervention to be clinically applicable. Thus, we evaluated 2.2. Diopter and ocular axial length measurements
the effects of whole-body UVA irradiation plus oral administration
of riboflavin on the biochemical and biomechanical properties of After general anesthesia by injecting ketamine hydrochloride
sclera in myopic guinea pigs to develop a unique, non-invasive, and into the thigh muscles, the axial length was measured with A-Node
practical therapy to control the progression of myopia. This is the Ultrasonic Diagnostic Equipment (Meda Co., Ltd, China). Retinos-
first report to investigate the effect of whole-body UVA irradiation copy optometry (Fizz Kang Visual Light Company, China) was used
and riboflavin without surgical exposure of the sclera in myopic to examine the diopter after cycloplegic pre- and post-experiments.
eyes.
2.3. Biomechanical measurements

2. Materials and methods Guinea pigs were killed after two weeks and eyes were put into
preservation solution and cryopreserved in liquid nitrogen after
2.1. Animals and treatments enucleation, after which the eyes were rewarmed rapidly in a
37e38  C water bath. Stressestrain tests were performed for sclera
All animal experiments conformed to “Regulations of Animal specimens using a microcomputer-controlled biomaterial tester
Experiments in Zhejiang Province” and were approved by Zhejiang (Instron Test Equipment Trade, Ltd, USA). The 15 mm  4 mm
Chinese Medical University. A total of 30 guinea pigs (all females, scleral strips were clamped horizontally with a stress level of
four-weeks old) were treated with a 10.0 D concave lens on the 0.005e0.04 N, and the strain was increased linearly at a velocity of
right eye for two weeks to establish the myopic model. The concave 5 mm/min to test the tensile strength until the specimen broke. The
lens was a 10.00 D poly(methyl methacrylate) (PMMA) lens ultimate load, ultimate stress, and ultimate strain were recorded
(Shanghai Freshkon Contact Lenses Co., Ltd., China) with a diameter using the microcomputer-controlled tester over time for further
of 11.00 mm and inner arc curvature of 9.00 mm (Fig. 1). All guinea analysis.
pigs were randomly divided into five groups (n ¼ 6 in each):
group A (vitamin C þ whole-body fluorescent lamp irradiation, 2.4. Posterior sclera thickness
control group), group B (vitamin C þ whole-body UVA irradiation,
control group), group C (vitamin C þ riboflavin þ whole-body After removing the vitreous, retina, and surrounding connective
fluorescent lamp irradiation, control group), group D (vitamin tissue, scleral tissue (diameter, 6 mm) taken from the posterior
C þ riboflavin þ whole-body UVA irradiation, experimental group), sclera was immediately immersed in 4% paraformaldehyde to be
and group E (riboflavin þ whole-body UVA irradiation, experi- fixed for 24 h, after which the sample was dehydrated and paraffin-
mental group). As mentioned above, all guinea pigs were fed a basal embedded to make a 4-mm section of the sclera; thin 4-mm sections
diet and 100 mg of vitamin C or 0.4 mg riboflavin in 0.1% solution by were stained with hematoxylin and eosin and periodic acid-Schiff
gavage three times per day from three days before the beginning of (PAS). A Zeiss light microscope (Axiomat, Zeiss, Oberkochen, Ger-
many) was used to measure the specimens (scleral tissue thickness
in the central region) at various magnifications.

2.5. Western blotting

Tissue homogenate was produced from 3 mm  4 mm sections


of posterior scleral tissue (100 mg) with 0.5e1 mL of cold lysis
buffer, after which it was centrifuged at 4  C and 10,000 rpm for
5 min. The protein concentrations were determined using bovine
serum albumin (BSA) as standards based on the Bradford assay
(Bio-Rad, BioTek Instruments, Inc. USA). The protein solution was
then separated by 10% polyacrylamide gel electrophoresis with 5%
stacking gels with sodium dodecyl sulfate polyacrylamide gel
electrophoresis (SDS-PAGE). The separated proteins were trans-
ferred onto nitrocellulose membranes, after which the membranes
were washed in Tris-buffered saline with 0.1% Tween-20 (TBS-T)
and blocked in plates with diluted matrix metalloproteinase-2
(MMP-2) antibody BA0569 (Wuhan Boster Biological Technology,
Fig. 1. Guinea pig models of lens-induced myopia. A: A single-use Murphy dropper
was pruned to construct a frame and the concave lens was agglutinated to the self-
LTD., China) or tissue inhibitor of MMP-2 (TIMP-2) antibody BA0576
made spectacle. B: The self-made spectacle was sutured and fixed to the right eye of (Wuhan Boster Biological Technology, LTD., China) for 10e12 h at
the guinea pig. 4  C. Membranes were then washed three times in TBS-T and
X. Li et al. / Experimental Eye Research 165 (2017) 1e6 3

incubated with secondary antibodies for 1 h at 26  C, after which in the lens-induced eye measured 3.33 ± 0.17 D (D) and 3.27 ± 0.16
they were washed another three times in TBS-T. Protein blots on the D (E) at day 0 and 1.47 ± 0.13 D and 1.53 ± 0.22 D 14 days later,
membranes were then enhanced by chemiluminescence (ECL respectively. A comparison of the mean changes in diopter of the
Detection Kit, Beyotime, China) and exposed using the FluorChem E right eyes between control groups (A, B, C) and experimental
System. Films were scanned into digital images and the primary groups (D, E) yielded a lesser net increase for myopic eyes in the
densities of the band of MMP-2, TIMP-2, and glyceraldehyde-3- experimental groups (D, E) (A vs. D, A vs. E, B vs. D, B vs. E, C vs. D, C
phosphate dehydrogenase (GAPDH) were measured with Quan- vs. E, all p < 0.01) (Fig. 2B). There was no significant difference in the
tity One (Bio-Rad). axial length and diopter (p > 0.05) between groups D and E. These
results demonstrated that our intervention method may suppress
2.6. Transmission electron microscopy the development of myopia.

A small mark was made at the temporal corneal limbus using an 3.2. Posterior scleral thickness and biomechanical measurements
indelible ink marker pen to allow for orientation of the eye cup
after enucleation. Eyes were immersed in 0.05 M cacodylic acid The scleral thickness of the control eye was thicker than the
sodium and 2.5% glutaraldehyde with phosphate buffer saline (pH lens-induced eye in all groups, as indicated in Fig. 2C. In the
7.4) for 2 h. After this period, the cornea and lens were dissected, experimental groups (D, E), mean thickness of myopic eyes showed
leaving the mark in the limbal region. A cornea trephine (diameter, a significantly lower net decrease (A vs. D, A vs. E, B vs. D, B vs. E, C
6 mm) was then used to punch out a posterior tissue button vs. D, C vs. E, all p < 0.01) compared with myopic eyes in control
(containing retina, choroid, and sclera). Two 2 mm  1 mm strips of groups (A, B, C). Although no significant difference was observed
sclera tissue were excised from the tissue button near the nasal between experimental groups (D, E), our results indicated that
region, 1 mm from the optic nerve with razor blades. The strips systematic whole-body UVA irradiation with oral administration of
were immersed in 0.05 M cacodylic acid sodium and 2.5% glutar- riboflavin could thicken the sclera in myopic eyes.
aldehyde for 24 h. Tissue samples were post-fixed in 1% osmium Compared with control eyes, the ultimate load and stress
tetroxide with phosphate buffer saline (pH 7.4) for 1 h at 4  C, assessment were lower and the strain assessment was higher for
stained with 1% uranyl acetate with double distilled water for 2 h, myopic eyes in each group. Moreover, the ultimate load and stress
and then rinsed and dehydrated in graded acetone before embed- assessment for myopic eyes in the experimental groups were
ding in Araldite. Scleral samples were analyzed using transmission significantly higher and strain assessment for myopic eyes in the
electron microscopy. experimental groups was significantly lower than myopic eyes in
control groups (Fig. 2DeF). Fig. 2D showed that after 14 days of
2.7. Statistical analyses myopic modeling, myopic eyes in experimental groups (D, E) had
higher values of ultimate load than myopic eyes in control groups
Statistical analysis with a significance level of p < 0.05 was (A, B, C) (Fig. 2D, A vs. D, A vs. E, B vs. D, B vs. E, C vs. D, C vs. E, all
performed using Tukey's multiple comparisons test with SPSS p < 0.01); Fig. 2E showed that after 14 days of myopic modeling,
software (v17.0, SPSS Inc, USA). Data are presented as myopic eyes in experimental group D had higher values of stress
mean ± standard deviation (SD). assessment than myopic eyes in the control groups (A, B) (Fig. 2E, A
vs. D, B vs. D, all p < 0.01); Fig. 2F showed that after 14 days of
3. Results myopic modeling, myopic eyes in experimental groups (D, E) had
lower values of strain assessment than myopic eyes in control
3.1. Establishment of myopia models groups (A, B) (Fig. 2F, A vs. D, B vs. D, all p < 0.01; B vs. E, p < 0.05).
These results indicated that sclera in high myopic eyes was more
After two weeks, we compared pre-test and post-test parame- flexible and its load bearing capacity was lower than emmetropia,
ters of ocular axial length and diopter. The results showed that all but whole-body UVA irradiation with oral administration of ribo-
data in the lens-induced eye had significantly longer axial lengths flavin could increase the biomechanical characteristics of sclera in
and higher myopic diopter compared with data from the pre-test, myopic eyes.
which indicated that the myopic models were established
successfully. 3.3. Protein expression level of MMP-2 and tissue TIMP-2
As shown in Fig. 2, in control groups (A, B, C), mean axial length
in the lens-induced eye was 7.37 ± 0.25 mm (A), 7.23 ± 0.19 mm (B), MMP-2 and TIMP-2 play key roles in the development of
and 7.27 ± 0.24 mm (C) before establishing the myopic model, and myopia. Visual signals may modulate gene expression of selected
8.31 ± 0.09 mm, 8.25 ± 0.11 mm, and 8.11 ± 0.17 mm 14 days later, MMPs and TIMPs to control scleral remodeling, the mechanical
respectively. In experimental groups (D, E), which underwent oral properties of the sclera, axial elongation, and the refractive state
administration of riboflavin and whole-body UVA irradiation, mean (Siegwart and Norton, 2005). It is believed that the increased
axial length in the lens-induced eye were 7.09 ± 0.18 mm (D) and expression of MMP-2 and the decreased expression of TIMP-2
7.19 ± 0.11 mm (E) at day 0 and 7.94 ± 0.10 mm and 7.99 ± 0.09 mm promote myopia.
14 days later, respectively. The measurement of the axial lengths The results shown in Fig. 3 demonstrate that whole-body UVA
revealed a lower net increase in myopic eyes of experimental irradiation plus oral administration of riboflavin regulates the
groups (D, E) compared with myopic eyes in control groups (A, B, C) expression of MMP-2 and TIMP-2. In the control groups (A, B, C),
(Fig. 2A). In the present study, significant differences were observed myopic treatment significantly increased the expression of MMP-2
between the control and experimental groups (A vs. D, A vs. E, B vs. (Fig. 3B) and significantly decreased the expression of TIMP-2
D, B vs. E, all p < 0.01). However, no significant difference (p > 0.05) (Fig. 3C), whereas in the experimental groups (D, E), myopic
was observed between group C and experimental groups (D, E). treatment did not significantly increase the expression of MMP-2
In control groups (A, B, C), mean diopter in the lens-induced eye (Fig. 3B) and did not significantly decrease the expression of
measured 3.37 ± 0.29 D (A), 3.40 ± 0.28 D (B), and 3.51 ± 0.29 D (C) TIMP-2 (Fig. 3C), but actually increased the expression of TIMP-2.
at day 0 and 2.21 ± 0.25 D, 2.26 ± 0.26 D, and 2.03 ± 0.10 D 14 Overall, our results demonstrated that whole-body UVA irradia-
days later, respectively. In experimental groups (D, E), mean diopter tion plus oral administration of riboflavin could suppress myopia
4 X. Li et al. / Experimental Eye Research 165 (2017) 1e6

Fig. 2. Ocular axial length, diopter, posterior scleral thickness, ultimate load, stress assessment, and strain assessment data. A: Significant differences in axial length were observed
between groups in the lens-induced eye at 14 days (**D p < 0.01 vs. A, B, **E p < 0.01 vs. A, B). B: Comparison of the mean diopter of lens-induced eyes at 14 days yielded a
significantly lower value (**D p < 0.01 vs. A, B, C; **E p < 0.01 vs. A, B, C). C: Scleral thickness in the lens-induced eye: **D p < 0.01 vs. A, B, C; **E p < 0.01 vs. A, B, C. D: Ultimate load in
the lens-induced eye: **D p < 0.01 vs. A, B, C; **E p < 0.01 vs. A, B, C. E: Ultimate stress in the lens-induced eye: **D p < 0.01 vs. A, B. F: Ultimate strain in the lens-induced eye: **D p <
0.01 vs. A, B; *E p < 0.05 vs. B.

Fig. 3. Western blot analysis of MMP-2 and TIMP-2. (GAPDH: glyceraldehyde-3-phosphate dehydrogenase). A: Representative picture of Western blot analysis (WB). B: Statistical
summary of the expression of MMP-2 in WB. C: Statistical summary of the expression of TIMP-2 in WB. **: p < 0.01 vs. control.

progression. Compared with control groups (A, B, C), the density of scleral fi-
broblasts increased significantly in myopic eyes in the experimental
groups (D, E), and more fibroblasts were present in myopic eyes in
3.4. Transmission electron microscopy of scleral tissue group D and E (Fig. 4). In total, whole-body UVA irradiation plus
oral administration of riboflavin could enhance the hyperplasia of
Transmission electron microscopy was used to detect the hy- fibroblasts in scleral tissue.
perplasia of fibroblasts in scleral tissue 14 days after treatment.
X. Li et al. / Experimental Eye Research 165 (2017) 1e6 5

Fig. 4. Photomicrograph of the sclera in different groups demonstrating the hyperplasia of fibroblasts at 14 days after myopic treatment. Arrows indicate the aggregation and
cellular bundles of fibroblasts in the sclera.

4. Discussion specific proteins, such as TIMPs (Siegwart and Norton, 2002).


Biomechanical scleral changes in myopia include of a net loss of the
Myopia is often accompanied by decreased scleral collagen matrix induced by increases in levels of both latent and active
content, changes in collagen fiber diameter and distribution, and MMP-2 (Rada et al., 2006; Guggenheim and McBrien, 1996). These
decreased scleral thickness (McBrien et al., 2001). Riboflavin and results were confirmed in the present study, showing that the
UVA treatment, as well as other chemical cross-linking methods expression of MMP-2 increased and the expression of TIMP-2
such as glutaraldehyde injection (Charulatha and Rajaram, 2003), decreased in lens-induced eyes. Interestingly, myopic eyes in the
are rarely used to control myopia due to their potential debatable experimental groups showed a lower net increase in MMP-2 and a
roles in the sclera. The applications of these methods are limited lower net decrease of TIMP-2, suggesting that our method does not
due to the available area of irradiation, the requirement of surgical strongly affect the sclera. However, further studies are required to
exposure of the sclera (Wollensak and Iomdina, 2009), and the explore the underlying mechanisms.
potential cytotoxic risk to the corneal endothelium, lens, and retina Based on electron microscopy, the sclera consists of fibroblasts
at an inappropriate dosage (Wollensak et al. 2003, 2005; Zhao et al., and collagen bundles, which can increase strength through changes
2013). Our data suggest that systematic whole-body UVA irradia- in the biomechanical properties of the sclera. In myopic eyes in our
tion plus oral administration of riboflavin can effectively improve control groups (A, B, C), the diameter of the posterior sclera fibro-
the structural, biochemical, and biomechanical characteristics of blasts was thinner compared with that of myopic eyes in experi-
sclera, delaying myopic development in a guinea pig myopic model. mental groups (D, E), and there was vacuolar degeneration in the
Indeed, myopia-induced decreases in the diameter of collagen fibroblasts. A previous report showed that the major differences
fibrils and collagen fiber bundles result in elongation and localized were in morphology and in the diameter of fibroblasts in the pos-
ectasia of the sclera (Rada et al., 2006). Moreover, elongation of the terior sclera between normal and myopic eyes (Liu et al., 1986),
eyes is determined mechanistically by the biochemical and which agrees with our results.
biomechanical properties of the sclera (Rada et al., 2000). Common According to the above reports, riboflavin, no systemic side ef-
approaches to change the biochemical and biomechanical proper- fects, can increase the absorption of UVA to limit the damage of
ties of the sclera include surgical exposure of the sclera, which is UVA to ocular tissues under the appropriate doses during collagen
invasive and painful. In this study, we attempted to identify a novel cross-linking (Zhao et al., 2013). However, no reports have provided
approach to change the biochemical and biomechanical properties the exact dose of orally administered riboflavin, which is important
of the sclera. We showed that the oral administration of riboflavin to ensure that the concentration in the sclera is sufficient to have a
combined with whole-body UVA irradiation without surgical therapeutic effect.
exposure of the sclera and direct UVA irradiation on the sclera The effects of UVA on the retina were a concern during this
could suppress axial length elongation and myopic diopter incer- study. Thus, the parameters of UVA irradiation were carefully
asing in a guinea pig myopic model. selected. Wollensak and Iomdina (2009) confirmed that there were
Pathological changes in the sclera with progressive scleral no side effects on the retina or retinal pigment epithelium using
thinning and a decreased biomechanical extensibility are impor- UVA under 370 nm and 3 mW/cm2. Moreover, UVA was applied for
tant features and common causes of severe myopia (McBrien and collagen cross-linking at 370 nm, where riboflavin absorption
Gentle, 2003; Rada et al., 2006). Choi et al. suggested that the peaks and there is no evidence of tissue damage in this band (Zhao
collagen cross-linking technique induced by riboflavin and UVA et al., 2013). Therefore, in this study, UVA at 370 nm and 3 mW/cm2
leads to increases in the density, area, diameter, and thickness of were selected and UVA irradiation was applied using a whole-body
human scleral tissues (Choi et al., 2013). In the present study, the irradiation approach.
experimental groups with oral riboflavin and whole-body UVA Vitamin C, unlike other antioxidants, does not absorb radiation
irradiation showed a lower net decrease in the development of in the UVA range (Ou-Yang et al., 2004) and can slow the absorption
myopia, especially in scleral thickness and ultimate load, suggest- and degradation of riboflavin (Lee et al., 1998). Moreover, vitamin C
ing that our method may be associated with scleral collagen cross- can counteract DNA damage and mutagenesis induced by the UVA
linking in the myopic eye. and riboflavin (Besaratinia et al., 2007). Therefore, the possibility of
During the development of myopia, the sclera undergoes active vitamin C to interfere with the optical effects of UVA and riboflavin
remodeling that involves the regulation of MMPs and TIMPs (Kim could be ruled out. The protective effects of vitamin C observed in
et al., 2001; Shelton and Rada, 2007). Therefore, the expression this study may be ascribed to its antioxidant and free radical
variations of these two proteins among the groups were evaluated scavenging capacities (Catani et al., 2005).
in our myopic model. Increased scleral MMP-2 expression in form- In summary, we present the first in vivo study on decreasing
deprivation myopia has been observed in guinea pigs at the protein axial length elongation and limiting increases in myopic diopter in
level (Wang et al., 2010). Meanwhile, MMP-2 can be controlled by a guinea pig myopic model using non-invasive oral administration
numerous factors including down-regulation of the expression of of riboflavin combined with whole-body UVA irradiation. The
6 X. Li et al. / Experimental Eye Research 165 (2017) 1e6

above combinative intervention led to significant changes in pro- 1514e1521.


Lee, K.H., Jung, M.Y., Kim, S.Y., 1998. Effects of ascorbic acid on the light-induced
tein levels of MMP-2 and TIMP (related to scleral remodeling) and
riboflavin degradation and color changes in milks. J. Agric. Food. Chem. 46,
biomechanical rigidity of the sclera in the myopic model mentioned 407e410.
above. In our study, we found no significant difference in the axial Liu, K.R., Chen, M.S., Ko, L.S., 1986. Electron microscopic studies of the sclera
length, stress assessment and strain assessment between group C collagen fiber in excessively high myopia. Taiwan. yi. xue. hui. za. Zhi 85,
1032e1038.
with experimental groups (D, E). It is known that the wavelength of Lu, Q., Zheng, Y., Sun, B., Cui, T., Congdon, N., Hu, A., Chen, J., Shi, J., 2009.
fluorescent lamps range from 400 to 700 nm (Pasquale et al., 2017; A population-based study of visual impairment among pre-school children in
Karin et al., 2017). Within this field, some wavelengths of light may Beijing: the Beijing study of visual impairment in children. Am. J. Ophthalmol.
147, 1075e1081.
have an effect, similar to UVA (370 ± 5 nm), to delay myopic McBrien, N.A., Cornell, L.M., Gentle, A., 2001. Structural and ultrastructural changes
development with the oral administration of riboflavin. Further to the sclera in a mammalian model of high myopia. Invest. Ophthalmol. Vis.
systemic studies are required to determine the underlying mech- Sci. 42, 2179e2187.
McBrien, N.A., Gentle, A., 2003. Role of the sclera in the development and patho-
anisms, address the potential toxicity (such as impairment of the logical complications of myopia. Prog. Retin. Eye. Res. 22, 307e338.
retina), and develop protective solutions (such as protection of the Ou-Yang, H., Stamatas, G., Saliou, C., Kollias, N., 2004. A chemiluminescence study of
retina). UVA-induced oxidative stress in human skin in vivo. J. Invest. Dermatol 122,
1020e1029.
Pasquale, O., Luigi, G., Massimiliano, M., Stefano, O., 2017. The photoluminescence of
Conflict of interest a fluorescent lamp: didactic experiments on the exponential decay. Phys. Educ.
52, 015011.
Phillips, J.R., Khalaj, M., McBrien, N.A., 2000. Induced myopia associated with
The authors in this article have no potential conflicts of interest.
increased scleral creep in chick and tree shrew eyes. Invest. Ophthalmol. Vis.
Sci. 41, 2028e2034.
Disclosure Rada, J.A., Achen, V.R., Penugonda, S., Schmidt, R.W., Mount, B.A., 2000. Proteo-
glycan composition in the human sclera during growth and aging. Invest.
Ophthalmol. Vis. Sci. 41, 1639e1648.
Contributions of authors: Designing the study (X.X.L., M.Q.W.); Rada, J.A., Shelton, S., Norton, T.T., 2006. The sclera and myopia. Exp. Eye. Res. 82,
conducting the study (H.L., L.Z.); collecting and managing the data 185e200.
(X.X.L., H.L., L.Z.); analyzing and interpreting the data (X.X.L., L.Y.Z., Resnikoff, S., Pascolini, D., Mariotti, S.P., Pokharel, G.P., 2008. Global magnitude of
visual impairment caused by uncorrected refractive errors in 2004. Bull. World.
J.J.H.); preparing, reviewing, and approving the manuscript (X.X.L., Health. Organ 86, 63e70.
M.Q.W.). Robaei, D., Huynh, S.C., Kifley, A., Mitchell, P., 2006. Correctable and non-correctable
visual impairment in a population-based sample of 12-year-old Australian
children. Am. J. Ophthalmol. 142, 112e118.
Acknowledgments Shelton, L., Rada, J.S., 2007. Effects of cyclic mechanical stretch on extracellular
matrix synthesis by human scleral fibroblasts. Exp. Eye. Res. 84, 314e322.
We thank 100Biotech Co. Ltd for assistance of performing pretest Siegwart Jr., J.T., Norton, T.T., 2002. The time course of changes in mRNA levels in
tree shrew sclera during induced myopia and recovery. Invest. Ophthalmol. Vis.
and affording the labs. This Project was supported by Zhejiang Sci. 43, 2067e2075.
Province Science and Technology Department (2014C03042-1). Siegwart Jr., J.T., Norton, T.T., 2005. Selective regulation of MMP and TIMP mRNA
levels in tree shrew sclera during minus lens compensation and recovery.
Invest. Ophthalmol. Vis. Sci. 46, 3484e3492.
References
Vitale, S., Ellwein, L., Cotch, M.F., Ferris 3rd, F.L., Sperduto, R., 2008. Prevalence of
refractive error in the United States, 1999e2004. Arch. Ophthalmol. 126,
Besaratinia, A., Kim, S.I., Bates, S.E., Pfeifer, G.P., 2007. Riboflavin activated by ul- 1111e1119.
traviolet A1 irradiation induces oxidative DNA damage-mediated mutations Wang, S.R., Ye, J.J., Long, Q., 2010. Expressions of collagen I, matrix
inhibited by vitamin C. Proc. Natl. Acad. Sci. U. S. A. 104, 5953e5958. metalloproeinases-2, and tissue inhibitor of matrix metalloproteinase-2 in the
Catani, M.V., Savini, I., Rossi, A., Melino, G., Avigliano, L., 2005. Biological role of posterior sclera of newborn Guinea pigs with negative lens-defocused myopia.
vitamin C in keratinocytes. Nutr. Rev. 63, 81e90. Zhongguo. Yi. Xue. Ke. Xue. Yuan. Xue. Bao 32, 55e59.
Charulatha, V., Rajaram, A., 2003. Influence of different crosslinking treatments on Wollensak, G., Iomdina, E., 2009. Long-term biomechanical properties of rabbit
the physical properties of collagen membranes. Biomaterials 24, 759e767. sclera after collagen crosslinking using riboflavin and ultraviolet A (UVA). Acta.
Choi, S., Lee, S.C., Lee, H.J., Cheong, Y., Jung, G.B., Jin, K.H., Park, H.K., 2013. Structural Ophthalmol. 87, 193e198.
response of human corneal and scleral tissues to collagen cross-linking treat- Wollensak, G., Iomdina, E., Dittert, D.D., Salamatina, O., Stoltenburg, G., 2005. Cross-
ment with riboflavin and ultraviolet A light. Lasers. Med. Sci. 28, 1289e1296. linking of scleral collagen in the rabbit using riboflavin and UVA. Acta. Oph-
Guggenheim, J.A., McBrien, N.A., 1996. Form-deprivation myopia induces activation thalmol. Scand. 83, 477e482.
of scleral matrix metalloproteinase-2 in tree shrew. Invest. Ophthalmol. Vis. Sci. Wollensak, G., Spoerl, E., 2004. Collagen crosslinking of human and porcine sclera.
37, 1380e1395. J. Cataract. Refract. Surg. 30, 689e695.
Holden, B., Sankaridurg, P., Smith, E., Aller, T., Jong, M., He, M., 2014. Myopia, an Wollensak, G., Spoerl, E., Wilsch, M., Seiler, T., 2003. Endothelial cell damage after
underrated global challenge to vision: where the current data takes us on riboflavin-ultraviolet-A treatment in the rabbit. J. Cataract. Refract. Surg. 29,
myopia control. Eye (Lond) 28, 142e146. 1786e1790.
Karin, K., Takayuki, T., Mark, A.S., 2017. Performance of arabidopsis thaliana under Zhao, X., Wang, M.M., Zhang, F.J., 2013. Advanced research in the safety of UVA-
different light qualities: comparison of light-emitting diodes to fluorescent riboflavin cross-linking for strengthening corneal and scleral tissue. Zhong-
lamp. Funct. plant Biol. 1445e4408. hua. Yan. Ke. Za. Zhi 49, 189e192.
Kim, J.W., Lindsey, J.D., Wang, N., Weinreb, R.N., 2001. Increased human scleral
permeability with prostaglandin exposure. Invest. Ophthalmol. Vis. Sci. 42,