Low Vitamin E Level as a Subliminal Risk Factor in a Rat Model of Prednisolone-Induced Cataract

Abstract

purpose. To investigate the relationship between vitamin E deficiency and prednisolone-induced cataract formation, long-term examination of lens changes was performed in rats under the condition of vitamin E deficiency or supplementation and administration of prednisolone.

methods. Rats were divided into six groups: normal chow (N), vitamin E–deficient chow (ED), normal chow with prednisolone instillation (NP), vitamin E–deficient chow with prednisolone instillation (EDP), NP treatment with vitamin E supplementation (NP+VE), and EDP treatment with vitamin E supplementation (EDP+VE). Prednisolone (1 mg/kg · d) and vitamin E (5%; 10 μL per administration per eye, 1 mg/kg · d) were applied in the cul-de-sac. Lens changes were documented and analyzed. Vitamin E status was confirmed by measuring peroxide-induced hemolysis.

results. After 15 months, 91.7% of the eyes in the EDP group showed development of anterior and posterior cortical cataracts. Supplementation with vitamin E significantly reduced cataract formation (to 38.9% of eyes). Neither a vitamin E–deficient diet nor prednisolone treatment alone significantly increased cataract formation. Hemolysis-susceptibility tests confirmed the expected vitamin E status of rats fed vitamin E–deficient chow and rats supplemented with eye drops containing vitamin E.

conclusions. Vitamin E deficiency and long-term prednisolone treatment together cause cataracts. Singly, however, both conditions are subliminal cataractogenic risk factors.

Since Black et al. ¹ reported a high incidence of posterior subcapsular cataracts in 44 patients with rheumatoid arthritis in 1960, numerous studies concerning steroid-induced posterior subcapsular cataract have been performed. A review of corticosteroid-induced cataract was published by Urban and Cotlier ² in 1986. The incidence of steroid cataract in patients receiving renal transplants varied from 6.5% to 96% because of differences in the duration of follow-up, patient age, and methods of steroid administration. ³ However, there is no doubt that long-term steroid application is a risk factor for cataract formation. 

There is no adequate therapy for steroid-induced cataract except cataract surgery. The search for treatments to prevent steroid-induced cataract is hampered by the lack of a suitable animal model. We have developed a prednisolone-induced cataract model using rats physically compromised by 2-Gy x-ray irradiation of the eye and subsequent long-term prednisolone administration. ^{4 5} In these studies, either topical (eye drops) or systemic administration of prednisolone acetate induces morphologic changes in the rat lens similar to those found in human steroid-induced cataracts. ⁵ Treatment of these animals with topical vitamin E ophthalmic solution prevents opacification. ⁶ Based on these results, studies were initiated to determine whether cataracts could be induced in rats by combining a low-vitamin-E diet and long-term prednisolone administration. Cortical cataracts developed in animals treated in this manner. Application of topical vitamin E reduced the incidence of these cataracts. 

Materials and Methods

Animals were cared for and handled in accordance with the Guidelines for Animal Experiments in Kanazawa Medical University and the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. 

Seven-week-old male Brown-Norway rats were obtained from Sankyo Labo Service Co., Inc. (Toyama, Japan). The animals were maintained on normal rat chow (Labo MR stock; Sankyo Labo Service Co., Inc.) for 1 week before the start of the experiment. The anterior segments of all rats were examined using a slit lamp microscope (SL-5D; Topcon, Tokyo, Japan) and 53 rats with normal-appearing eyes were selected. Baseline values were recorded using an anterior eye segment analysis system (EAS-1000; Nidek, Gamagori, Japan) before starting the experiment. ⁷  

Treatment Groups

The rats were divided into six groups. Six rats were fed a normal diet and served as the control (N group). The right eye of the control rats was treated with the vehicle used to prepare the vitamin E ophthalmic solution. Six rats were fed vitamin E–deficient chow (ED group). Both eyes of these animals were treated with the vehicle used to prepare the vitamin E ophthalmic solution. Twelve rats were fed normal chow and had 1% prednisolone acetate ophthalmic solution instilled in both eyes (NP group). Another 12 rats received the same treatment as the NP group and were fed vitamin E–deficient chow (EDP group). The fifth group (n = 8) received the same treatment as the NP group, but the rats were also treated with 5% vitamin E ophthalmic solution (VE) administered daily into both eyes (NP+VE). The final group (n = 9) received the same treatment as the EDP group, but with the daily administration of 5% vitamin E in both eyes (EDP+VE). The study was performed in a single-masked manner so that drug administration, documentation of lens conditions, and image analyses were conducted by three different examiners. Normal rat chow and vitamin E–deficient chow (AIN76) were provided by Sankyo Labo Service Co., Inc. Prednisolone acetate, vitamin E ophthalmic solution, and vitamin E vehicle were obtained from Santen Pharmaceutical Co., Ltd. (Osaka, Japan). 

Drug Administration

Prednisolone acetate was dispersed in phosphate-buffered saline containing polysorbate 80. Ten microliters of a 1% solution of prednisolone acetate (ophthalmic solution, pH 7.3, ratio of osmolarity: 1.0:1.1) was instilled into the cul-de-sac of both eyes two times each day at 30-minute intervals (approximately 1 mg/kg · d). To avoid weakening the rats, the dose of prednisolone was sometimes gradually reduced when the health of the rats deteriorated. The normal dose of prednisolone acetate was resumed after the recovery of the rats, as described in previous reports. ^{4 5} Ten microliters of a 5% vitamin E ophthalmic solution (an emulsion of d-α-tocopherol acetate, pH 6.5, ratio of osmolarity: 1.0:1.0, in polysorbate 80) was administered into the cul-de-sac of both eyes two times each day at 30-minute intervals. Prednisolone administration was performed in the morning (11 AM) and vitamin E was instilled in the afternoon (4 PM). 

Documentation of Changes in Lens Transparency and Cataract Formation

Lens changes were evaluated using a slit lamp microscope and were documented once a month with the analysis system (EAS-1000; Nidek) in rats under maximum mydriasis (1% Nitten; atropine sulfate ophthalmic solution; Nihon Tenganyaku Kenkyusho Co., Ltd., Nagoya, Japan). Changes in lens transparency were objectively evaluated using the quantitative retroillumination capabilities of the analysis system, as previously described. ^{4 5 6}  

In brief, slit images were masked with a circle that excluded the cornea or iris. Reference images were obtained from each eye at the beginning of the experiment. Images obtained during the experiment were compared to these reference images. Changes in lens transparency were measured by determining the number of pixels in the image that showed increased opacity and compared to the reference image. Cataract diagnosis was performed using the slit lamp and from images generated using the analysis system. Human clinical criteria were applied in describing the location of opacities. 

Evaluation of General Health and Vitamin E Status

The body weight of the rats was checked once a week as an index of general health condition. The appearance of their skin and hair and their activity were also checked weekly (not reported). Previous reports demonstrated a relationship between vitamin E status and the ability of erythrocytes to be lysed by dialuric acid or hydrogen peroxide. ^{8 9 10 11} We therefore used hemolysis as an index of the vitamin E status of rats maintained on a vitamin E–deficient diet or supplemented with the vitamin. Blood (0.4 mL) was taken from the tail vein with a heparin-coated capillary and 6 mL citrate saline was added to 0.3 mL of the sample and centrifuged (2000 rpm, 10 minutes in a clinical centrifuge) and the pellet diluted to 2.5% red-cell suspension in phosphate-buffered saline. These cells were incubated at 37°C for 15 minutes and centrifuged and the pellet diluted in 0.9% saline to produce a 5% red-cell suspension. A portion of the suspension (250 μL) was put into three tubes (blank, control, and test sample). Phosphate buffer (250 μL) was added to the blank tube, and the same volume of 2.5% H₂O₂ was added to the control and test samples. To improve the standardization of the H₂O₂ decomposition rate, H₂O₂ was quickly added to the blood sample, as described by Ikehata et al. ¹¹ These tubes were incubated at 37°C for 3 hours. After incubation, 5 mL 0.9% saline and phosphate-buffered saline solution (1:1 mixture) were added to the blank and test tubes, and 5 mL distilled water was added to the control tube. After centrifuging, the optical density of the supernatant was measured at 540 nm. The optical density of the blank tube was subtracted from that of the control and the test sample tubes, and the percentage of hemolysis was calculated. 

Statistical Analysis

Experimental data were evaluated using Student’s two-tailed t-test. P < 0.05 was considered significant. 

Results

Body Weight Changes

The body weights of the rats from each group were similar until 4 to 5 weeks after the start of the experiment. After that, the order of mean body weight of each group was: ED > EDP+VE > N > EDP > NP+VE > NP. The trend of the mean body weight order did not change throughout the experiment (Fig. 1) . 

Lens Changes Observed by Slit Lamp and Analysis System Documentation

Most of the rats in the normal diet groups (N, NP, NP+VE) and in two of the groups fed vitamin E–deficient chow (ED and EDP+VE) showed no observable lens opacities throughout the experiment. All groups showed an increase in eye size and slight increases in lens nuclear opacity. In contrast, most of the animals that were fed vitamin E–deficient chow and were treated with prednisolone (EDP) showed development of anterior and posterior cortical cataracts and an opacified subcapsular layer (Fig. 2) . 

A few rats from each group, except those fed normal chow and not treated with prednisolone (N), showed slight lens changes. Two eyes from two animals of the ED group (16.7%) had small anterior cortical cataracts that developed 8 months after the start of the experiment. These crescent-shaped opacities elongated toward the equatorial region and had a filament-like extension by the end of the experiment. In four eyes of three animals (16.7%) from the NP group, crescent-shaped cataracts developed in the cortical layer at the equatorial region 10 months after the start of the experiment. Thereafter, these opacities elongated in the anterior and posterior directions. Rats from the NP+VE group (four eyes from three animals, 25%) also had opacities that were similar to those just described. 

In contrast to the occasional cataracts in the groups fed vitamin E–deficient chow or treated with prednisolone, 22 (91.7%) eyes of 12 animals of the EDP group had anterior cortical cataracts that developed 8 months after the start of the experiment. These opacities were originally similar to those seen in the ED group. By 15 months, however, animals in the EDP group had more severe anterior cortical cataracts and also had posterior cortical cataracts that included the subcapsular layer (Fig. 2) . 

When animals fed vitamin E–deficient chow and treated with prednisolone (EDP) were treated with topical vitamin E eye drops (EDP+VE), the incidence and severity of cataracts was significantly reduced (from 92% to 39%, P < 0.001; Fig. 2 ). 

It was also noted that the rats in the EDP group had a deeper anterior chamber than did the other groups (Figs. 2) . Inspection of these eyes showed that the increase in anterior chamber depth was associated with thinning of the cornea, increased corneal curvature and, perhaps, a decrease in the size of the lens. Because this result was not seen in any other group, it must have been due to the combined effect of vitamin E deficiency and corticosteroid treatment. 

Comparison of Light-Scattering Areas in Each Group

Figure 3 compares the mean area of increased light scattering in the treatment groups. The area of light scattering (as measured by the number of pixels with increased density in Scheimpflug slit images) increased steadily in all groups throughout the experiment. The light scattering in the EDP group, in particular, was significantly higher than that in the N group, from the third month until the end of the experiment. The ability of topical vitamin E to reduce light scattering after prednisolone administration (NP) or vitamin E deficiency plus prednisolone administration (EDP) was evaluated by comparing the densities among NP, NP+VE, EDP, and EDP+VE. The area of increased density in EDP+VE lenses was significantly smaller than that of the EDP lenses, from 2 months after the start of the experiment to the end of the study (P < 0.05). A significant difference in light scattering between NP and NP+VE was seen at 4, 5, 8, 13, and 15 months (P < 0.05). 

Evaluation of Vitamin E Status

Hemolysis induced by oxidative stress was used to evaluate the vitamin E status of each treatment group. All groups not fed a vitamin E–deficient diet showed low hemolysis (Fig. 4) . Animals fed vitamin E–deficient chow and not treated with supplemental vitamin E (ED and EDP) showed high hemolysis rates. In contrast, when animals in the EDP group were treated with supplemental vitamin E, hemolysis was reduced to the range seen in animals fed a normal diet. These results show that the vitamin E eye drops used in this study restored animals deprived of dietary vitamin E to normal systemic vitamin E status. 

Discussion

The pathogenesis of most human cataracts is believed to be the result of multiple factors acting over many years. Mechanisms of syn- and cocataractogenesis ¹² explain cataract formation due to an accumulation of several cataract risk factors. Syncataractogenesis represents the combination of two (or more) subliminal damaging factors that, only in combination, lead to lens opacities. In cocataractogenesis the direct cataractogenic effect of a substance is promoted when it is in combination with a subliminal factor that, on its own, has no effect. ¹² The data presented suggest that steroid-induced cataract is an example of multifactorial pathogenesis. 

The purpose of this experiment was to clarify the relationship between vitamin E in the diet and the development of prednisolone-induced cataract in rats. In our previous studies we found that low-dose x-ray irradiation (2 Gy) alone did not cause cataracts, but induced cataract formation when combined with prednisolone eye drops. ^{4 5 6} That the cataractogenic effect of the combination of x-rays and prednisolone is prevented by supplementation with vitamin E eye drops is interesting. ⁶ In the present study, we tested whether vitamin E deficiency would serve as a subliminal risk factor for cataract formation instead of 2-Gy x-ray irradiation. More than 80% of the rats in the vitamin E–deficient group showed clinically transparent lenses over the experimental term (Table 1) . This supports the view that, by itself, vitamin E deficiency is not cataractogenic, but has a subliminal risk potential. 

There are several reports concerning systemic vitamin E levels and the risk of cataract. A positive correlation has been reported between low vitamin E levels and cataract prevalence, ^{13 14 15} but a negative correlation (including oral vitamin E supplementation) has also been reported. ^{16 17 18} Despite many clinical, epidemiologic, and experimental studies concerning vitamin E levels in the body and cataract development, this problem remains unsolved, and the effects of long-term vitamin E administration on the inhibition of cataract development or delay of cataract progression are not well defined. 

The data provided in the present study suggest that vitamin E deficiency could increase the risk of cataract formation and that elevated levels of vitamin E may protect against other cataractogenic insults. Stephens et al. ¹⁹ reported that little change was seen in vitamin E concentrations in lenses of rats fed deficient or supplemented diets for 102 days, using gas chromatography-mass spectrometry (GC-MS) for precise measurement of vitamin E levels. In contrast, the investigators performed a study on vitamin E penetration of the eye and confirmed that 1% deuterium-labeled α-tocopherol acetate by eye drop application penetrates the aqueous humor and lens. ²⁰ It is known that drug penetration of the eye is more effective when the drug is instilled rather than administered to the whole body, because the eyeball becomes the drug’s target. This may explain why eye drop application of vitamin E supplementation showed positive effects. 

The mechanisms of corticosteroid-induced cataract formation have been speculated to be inhibition of the Na,K-ATPase pump, binding of corticosteroids to lens proteins and the subsequent formation of lysine-ketosteroid adducts, inhibition of glucose-6-phosphate dehydrogenase, loss of adenosine triphosphate (ATP), and secondary oxidation of -SH protein groups in lysine-ketosteroid adducts leading to the aggregation of crystallins. ^{2 21 22 23 24} There are two ways of thinking about steroid-induced cataract mechanisms: The steroid itself directly affects the lens, or metabolite(s) of the steroid secondarily affect the lens. Circumstantial evidence for a direct effect of the steroid on the lens is that lenses cultured in the presence of prednisolone-containing medium show opacity in the outer lens cortex. ^{21 24} This means that prednisolone can penetrate the lens and produce toxic effects directly in the lens. In our previous study, the application of eye drops resulted in a higher concentration of prednisolone in the lens than was obtained after retrobulbar or intramuscular injections. ²⁵ In addition, in vivo experimental results showed that the incidence of cataract formation was higher and the time course of development was faster with eye drop prednisolone application than in groups with systemic prednisolone application. ^{4 5} These results indicate that prednisolone-induced cataract formation is related to the prednisolone concentration in the lens and that the prednisolone itself may have direct influence on opacification. ⁵  

Nishigori et al. ^{26 27} found that steroid injected into a chicken egg produced cataract in 15-day-old embryos. On the basis of these results they suggested that the effect of the steroid is indirect. In these studies, hydrocortisone treatment increased the content of glucose ²⁶ and lipid peroxide ²⁷ in the lens after 24 to 48 hours. Lenses from treated embryos also had decreased antioxidant protective systems, such as catalase, glutathione peroxidase, aniline hydroxylase, superoxide dismutase, and glutathione reductase. ²⁸ Because we did not measure any biochemical parameters in this experiment, we must perform further experiments to determine the indirect effect of prednisolone treatment in our model. We have only experience in measuring blood glucose in the model of 2-Gy x-ray and prednisolone administration. Although the groups treated with intravenous prednisolone (including the vitamin E–treated group) showed a higher blood glucose level from the 11th to the 30th weeks compared with the control group (P < 0.05), the difference was slight and not in the range of that in diabetes. ⁶  

Vitamin E is well known to have functions including those of a chain-breaking antioxidant ²⁴ and membrane stabilizer. ^{29 30} Possible mechanisms of prednisolone-induced cataract under the condition of vitamin E deficiency are as follows: A vitamin E–deficient condition is very sensitive to oxidation stress. An accumulation of prednisolone (or metabolites of prednisolone) in the lens can cause oxidation of the lens fiber, leading to cataract. 

The mechanism of the effect of vitamin E against prednisolone acetate–induced cataract is probably its antioxidant effect and also its stabilization of the lens fiber membrane. 

Although it is well known that steroids induce posterior subcapsular cataracts in humans, our rat prednisolone acetate–induced cataract model first showed anterior cortical cataract, with posterior subcapsular cataract developing several months later. In the future it will be important to clarify the reasons for the differences between human steroid cataracts and experimentally induced prednisolone acetate cataracts in rats. 

General health conditions were monitored by changes in body weight. Our unexpected results showed that animals fed on the vitamin E–deficient chow had the highest mean body weight. This result may have been due to differences in the caloric content of the two diets or the amount of food consumed by animals fed the different formulations. 

In this investigation we used a type of area densitometry of slit images to document changes in light scattering in the anterior segment. The accuracy and reproducibility of data obtained in this manner were satisfactory, even in small experimental animals. Tests of reproducibility had a coefficient variation of 6%. ⁷ Although image analysis of the lenses of young animals is always complicated by lens (and eye) growth, the methodology used in this study provides reliable measurements for lens transparency changes or opacification during most of the life of the animal. 

²  MK and YBS contributed equally to the work.

 Submitted for publication April 9, 2001; revised October 29, 2001; accepted November 27, 2001.

 Commercial relationships policy: N.

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Mean values of body weight of the six experimental groups over the course of the experiment. N, 6 rats; ED, 6 rats; NP, 12 rats; EDP, 12 rats; NP+VE, 8 rats; EDP+VE, 8 rats. Significant body weight differences were found during the following terms: N vs. ED (4 to 18 weeks and 29 to 41 weeks); ED vs. EDP (4 to 47 weeks); N vs. NP (4 to 75 weeks); ED vs. EDP+VE (9 to 18 weeks); EDP vs. EDP+VE (6 to 8 weeks and 63 to 75 weeks); N vs. NP+VE (8 to 12 weeks and 27 to 58 weeks); NP vs. NP+VE (5 to 7 weeks).

Figure 2.

 

Representative timecourse of lens opacification in the EDP and the EDP+VE groups.

Figure 3.

 

Light scattering from lenses of the six experimental groups.

Figure 4.

 

Vitamin E status of animals in the six experimental groups. A higher percentage of hemolysis corresponded to lower vitamin E status. Bars, SD.

Table 1.

 

Type and Incidence of Cataract in All Groups at the End of the Experiment

Group (n)	Type of Opacity	Cataractous Eye Number/Eyes (%)
N (6)	No cataract	0
ED (6)	Ant. cortical	2/12 (16.7)
NP (12)	Ant./post. cortical	2/24 (16.7)
EDP (12)	Ant./post. cortical	22/24 (91.7)
NP+VE (8)	Ant./post. cortical	4/16 (25.0)
EDP+VE (9)	Ant./post. cortical	7/18 (38.9)

 Ant. cortical, anterior cortical cataract; Ant./post. cortical, anterior and posterior cortical cataract.

The authors thank David C. Beebe (Department of Ophthalmology and Visual Science, Washington University School of Medicine, St. Louis, MO) for valuable discussions and his kind revision of the manuscript.