Introduction

Age-related cataract is a major cause of visual impairment in the aging population worldwide. Age is the main risk factor of cataract, and it is sometimes assumed that cataract is simply an extension of the aging process [1]. The prevalence of this age-related disease is increasing dramatically as the proportion of the elderly in the population continues to rise. To date, a surgical procedure remains the only generally accepted treatment, whereas pharmacological prevention has been unsuccessful so far. Surgical cataract removal, however, is associated with some risks. In particular, it can lead to the development of early age-related macular degeneration (AMD), which is the most important cause of blindness among the elderly [2]. Estimates show that a delay of cataract development by 10 years reduces the need for a surgical intervention by 50 %. Therefore, creation of novel therapeutic and prophylactic modalities for cataract is necessary.

Animal research and observational studies suggest that antioxidant supplementation can decelerate aging of the eye and possibly provide some protection from cataract [3]. Indeed, oxidative stress is implicated in the aging process and in the pathogenesis of a wide range of age-related disorders including cataract formation and progression. Nonetheless, randomized controlled trials showed that supplementation with antioxidants neither prevents nor decelerates progression of age-related cataract [4, 5]. Moreover, a recent study reported an increased risk of cataract in people with increased intake of vitamins C and E [6]. According to the above studies, there is no evidence that antioxidant supplementation can reduce preexisting signs of cataract.

On the other hand, recently, using senescence-accelerated OXYS rats as an animal model of senile cataract, we showed that the antioxidant 10-(6′-plastoquinonyl) decyltriphenylphosphonium (SkQ1) at nanomolar concentrations can not only prevent the development of cataract but also partially reverse clinical signs of incipient stages of the disease in the lens [7–9]. Studies show that SkQ1 has a therapeutic potential against other age-related eye diseases including AMD and glaucoma [7–9]. SkQ1 in the form of eye drops (brand name Visomitin) was approved for the treatment of dry eye syndrome. Currently, clinical evaluation of anticataract effects of this drug is being actively conducted, but the evidence of its effectiveness is still not convincing enough. We believe that persuasive evidence of clinical efficacy of SkQ1 drops can come from a randomized experimental controlled study, such as the present work. Therefore, the purpose of this work was histomorphological examination of the influence of SkQ1 eye drops on the development of cataract in OXYS rats. Furthermore, we evaluated the effect of SkQ1 on mRNA expression of the Cryaa and Cryab genes, which encode subunits of α-crystallin. α-Crystallin is a major lens protein that acts as a chaperone, and plays a central role in the maintenance of lens transparency and refractive properties [10]. Most of crystallins are downregulated and dysregulated during development of age-related cataract, both in humans [11] and in OXYS rats [12–14]. In OXYS rats at the age of 20 days, clinical manifestations of cataract are absent according to our observations [12, 15]. The first signs of cataract (in 20 % of OXYS rats) appear by age 1.5 months, and at age 3 months, the morbidity reaches 100 % with predominance of a mild form of cataractous changes [12, 15]. A mature cataract is detectable in more than a half of 12-month-old OXYS rats, whereas Wistar rats (control parent strain) of the same age do not show any signs of this pathology. To assess the effectiveness of SkQ1 eye drops, we used OXYS rats from 9 to 12 months of age, when clinical signs of severe cataract are known to develop.

Materials and methods

The ethics statement

All animal procedures were in compliance with the Association for Research in Vision and Ophthalmology Statement on the Use of Animals in Ophthalmic and Vision Research as well as the European Communities Council Directive No. 86/609/EES. All manipulations with animals were approved by the Institutional Review Board (session No. 9) of the Institute of Cytology and Genetics (ICG) of the Siberian Branch of the Russian Academy of Sciences (SB RAS), according to “The Guidelines for Manipulations with Experimental Animals” (the decree of the Presidium of the Russian Academy of Sciences of April 02, 1980, No. 12000–496).

Animals and the treatment

Male senescence-accelerated OXYS rats and age-matched male Wistar rats (control animals) were obtained from the Breeding Experimental Animal Laboratory of ICG, SB RAS (Novosibirsk, Russia). The OXYS rat strain was derived from the Wistar strain at ICG as described previously [16, 17] and registered in the Rat Genome Database (http://rgd.mcw.edu/). At the age of 4 weeks, the pups were weaned and housed in groups of five animals per cage (57 × 36 × 20 cm). They were kept under standard laboratory conditions (22 ± 2 °C, 60 % relative humidity, and natural light), provided with standard rodent feed, PK-120-1, Ltd. (Laboratorsnab, Russia), and given water ad libitum.

SkQ1 was synthesized at the Institute of Mitoengineering of Moscow State University (Moscow, Russia). To assess the effects of SkQ1, male OXYS rats were randomly assigned to one of two groups (n = 15 each): {1} control or {2} instillation of one drop of a 250 nM SkQ1 solution in 0.9 % NaCl daily from 9 to 12 months of age. In addition, 15 untreated Wistar rats served as another control group. In total, 45 animals were used in the study (OXYS control, OXYS experiment, and untreated Wistar).

Ophthalmoscopic examination

An ophthalmologist examined all animals twice: before and after SkQ1 treatment, at the age of 9 and 12 months respectively. All rats were examined under a Heine BETA 200 TL Direct Ophthalmoscope (Heine, Germany) after dilatation with 1 % tropicamide. Assessment of stages of cataract was carried out according to the Lens Opacities Classification System III (LOCS III) [18]: grades NO1–NO6 and NC1–NC6 (nuclear opalescence and nuclear color, rated using standard samples, 1 through 6), grades C1–C5 (according to standards of cortical cataracts), and grades P1–P5 (according to standards of posterior subcapsular cataracts). Grades NO1, NC1, P1, and C1 correspond to a transparent lens. NO2, C2, and P2 correspond to ratings 1–4 on the standard decimal scale AREDS and are estimated as the first stage of cataract; NO3–NO4, C3–C4, and P3 correspond to ratings 5–8 on the standard decimal scale, and are estimated as the second stage of cataract. Grades NO5–NO6, C5, and P5 mean intensive cortical or nuclear opacity of the lens (corresponding to ratings 9–10 on the standard decimal scale), and are estimated as the third stage of cataract. The degree of lens opacification was classified using the double-blind method.

Morphological examination using light microscopy and electron microscopy

Histological and morphometric analyses were conducted on lenses of untreated Wistar rats and of SkQ1 eye drop–treated and untreated OXYS rats (eight randomly selected lenses in each of the three groups of animals). The lens of an eye was excised, fixed in 4 % paraformaldehyde, and prepared in the DMEM culture medium, which was used as a buffer in all processing procedures. Then the lenses were cut identically into four pieces, which were washed 3 times in DMEM and postfixed in 1 % OsO4 (SPI, USA), dehydrated in ethanol and acetone, soaked in a 1:1 mixture of epon–araldite and acetone for 12 h, and embedded in an epon–araldite mixture (SPI, USA) according to the manufacturer’s instructions. Semithin sections for light microscopy were made on an ultramicrotome EM UC7 (Leica, Germany), and the regions for electron microscopy were selected there as well; the sections were stained with 1 % azure II. The images for light microscopic examination were acquired using an AxioScope and the AxioVision software (Zeiss, Thornwood, NY). Ultrathin sections were also prepared on an EM UC 7 ultramicrotome from the regions selected on semithin sections, and contrast was enhanced using uranyl acetate and lead citrate. The sections were examined under a JEM 1400 transmission electron microscope (Jeol, Ltd., Japan) supplied with a digital camera Veleta (Olympus SIS, Germany).

Assessment of gene expression of αA- and αB-crystallin

RNA isolation

Total cellular RNA was isolated from a rat lens using phenol–chloroform extraction as follows. A tissue sample was placed in water-saturated phenol (pH 7.0; 10 volumes per gram of tissue) with addition of 0.5 % SDS (5 volumes per gram of tissue); then the tissue was homogenized for 15 min. After addition of 0.1 volume of 2 M sodium acetate pH 4.2, the tissue was homogenized again, the mixture was incubated for 10 min, and then chloroform extraction was performed. The supernatant was mixed with 2.5 volumes of 96 % ethanol, and the mixture was kept at −70 °C for 2 h. After that, the mixture was centrifuged (15 min at 13,000 rpm on an Eppendorf 5414 centrifuge; Eppendorf, Germany), and the pellet was dried and dissolved in 25 μl of RNA-grade H2O. The amount of isolated RNA was assessed by means of electrophoresis of 1 μl of RNA in a 1 % agarose gel. The RNA concentration was quantified in each sample on a spectrophotometer at 260 nm using absorbance ratios A260/A280 and A260/A320. The RNA samples were stored at −70 °C. Traces of genomic DNA were removed by means of DNase I (Promega, USA) according to the vendor’s instructions. After that, we repeated RNA extraction with the phenol–chloroform mixture and pure chloroform, followed by precipitation with ethanol as described above.

Reverse transcription

RNA (10 μg) and random hexamer primers (0.5 μg) were mixed in 11 μl of PCR-grade water. After RNA denaturation (65 °C for 5 min) and primer annealing (37 °C for 5 min), the solution of reverse transcriptase (19 μl) was added. The final reaction mixture contained the buffer for reverse transcriptase (20 mM Tris–HCl pH 8.3, 10 mM DTT, 100 mM KCl, 5 mM MgCl2), 500 μM deoxynucleotide triphosphates, and 40 U of MoMLV reverse transcriptase (Biosan, Novosibirsk, Russia). Synthesis of cDNA was performed at 37 °C for 1 h, 42 °C for 30 min, and 50 °C for 10 min. The polymerase was inactivated by heating this mixture at 75 °C for 5 min. For subsequent PCR, we used 0.25–0.50 μl of the resulting cDNA solution.

Preparation of standard cDNA

Aliquots (3 μl) from all cDNA samples were mixed, and the “average” solution was used to construct calibration curves, which were used for quantification of the relative cDNA levels of genes of interest, and of a reference gene in the cDNA samples.

Real-time PCR

Expression of the Cryaa and Cryab genes was determined using real-time PCR in the presence of the SYBR Green I dye (Molecular Probes, USA) on an iCycler iQ4 real-time PCR detection system (Bio-Rad Laboratories, USA). The housekeeping gene Rpl30 (encoding large ribosomal subunit protein 30) served as a reference gene. The following primer pairs were used: Rpl30, 5′-ATG GTG GCT GCA AAG AAG AC-3′ and 5′-CAA AGC TGG ACA GTT GTT GG-3′; Cryaa, 5′-AGC CGA CTG TTC GAC CAG TTC-3′ and 5′-AAC TTG TCC CGG TCA GAT CG-3′; and Cryab, 5′-CTT CGG AGA GCA CCT GTT GG-3′ and 5′-GAG AGA AGT GCT TCA CGT CCA-3′. The reaction mixture (final volume 20 μl) contained the standard PCR buffer (67 mM Tris–HCl pH 8.9, 16 mM [NH4]2SO4, 0.01 % Tween 20, 10 mM β-mercaptoethanol), 3 mM MgCl2, 0.2 mM dNTPs, SYBR Green I (1:20,000 dilution), 150 nM primers, and 0.4 U of Taq polymerase (ICG, Novosibirsk, Russia). The reaction was conducted under the following conditions: initial denaturation at 95 °C for 3 min, then five long cycles: denaturation at 95 °C for 30 s, annealing 30 s, elongation at 72 °C for 60 s; after that, 35 main cycles: denaturation at 95 °C for 20 s, annealing 20 s, elongation at 72 °C for 30 s, data collection based on fluorescence of the Rpl30 reaction at 84 °C for 30 s, and data collection based on fluorescence of the reactions of interest (Cryaa and Cryab at 87 °C) for 10 s. After completion of the PCR, the melting curves were recorded for specificity control. In each experiment, samples of a cDNA under study were mixed with primers for a gene of interest (fouir repeats per cDNA sample) in one multiwell plate, the same samples with primers for the reference gene (also four repeats), and “standard” cDNA diluted from 1:3 to 1:243 and mixed with the same primers (two to three repeats). For each cDNA sample, the PCR assay was repeated at least twice. The initial level of a cDNA in question was determined using standard calibration curves (versus standard cDNA), and the value obtained for each gene of interest was compared with the amount of the reference gene cDNA; thus, we determined differences in the expression of genes among the groups of rats [19]. The number of animals in each group was six or seven.

Western blotting

Protein concentrations were measured using a protein assay reagent (DC; Bio-Rad) with BSA as a standard. Equal amounts of protein (5 μg) were loaded in each lane. The blots were stained with Ponceau S to confirm equal protein loading. Proteins were separated in 12 % Tris-glycine gels (Novex; Invitrogen) and transferred to nitrocellulose membranes using standard methods. Each membrane was blocked with 5 % bovine serum albumin in Tris-buffered saline with 0.1 % Tween 20 (TBS-T buffer) for 1 h, and incubated overnight at 4 °C with a rabbit polyclonal antibody to αB-crystallin (1:1,000; Abcam, USA). The secondary antibody was a horseradish peroxidase–conjugated sheep anti-rabbit IgG antibody (1:10,000; Abcam, USA). After incubation with the secondary antibody, the chemiluminescent signal was measured and scanned, and emission intensity of the bands was quantified using the Image J software (NIH, Bethesda, MD, USA). β-Actin served as an internal loading control in each Western blot. The number of animals in each group varied from five to seven.

Statistical analysis

The design of the above experiments is a randomized experimental controlled study. Sampling of rats and eyes from each group for all types of experiments above was random; simple randomization was also used for distribution of the animals into the treated and untreated groups. All analyses in the Ophthalmoscopic examination section were double-blind (double-masked); the other experiments were not masked. The main variable was the cataract stage and the secondary variables were histomorphological parameters and the levels of α-crystallin expression. The data were analyzed using repeated-measures analysis of variance (ANOVA) and nonparametric tests using the statistical package Statistica 6.0 (StatSoft, USA). To assess the therapeutic effectiveness, we performed a dependent pairwise comparison of the cataract stages before and after treatment for every eye (t-test for dependent samples). One-way ANOVA was used to evaluate the differences between OXYS and Wistar rats and the effects of treatment with SkQ1 eye drops. The Newman–Keuls post hoc test was applied to significant main effects and interactions in order to assess the differences between some sets of means. The data were presented as mean ± SEM. The results were considered statistically significant if the p value was <0.05.

Results

Effects of SkQ1 treatment: ophthalmoscopic examination

Analysis of these data is shown in Fig. 1. Already during the first examination of OXYS rats (before treatment, age 9 months), we did not find a single eye that was free of pathological changes. All three types of cataract were detected in OXYS rats: posterior subcapsular (45 %), nuclear (38 %), and cortical (17 %). Pathological changes that were found during this first preliminary examination in one half of the eyes of untreated OXYS rats corresponded to the first stage of cataract, and in the other half to the second stage. According to the dependent pairwise comparison of the cataract stages at 9 and 12 months of age in the untreated OXYS rats, the cataract progressed (p < 0.00001), and the second examination of the eyes (age 12 months) revealed that 32 % of the eyes had the first stage of cataract, 62 % the second stage, and 4 % of the eyes reached the third stage.

Fig. 1
figure 1

SkQ1 eye drops inhibit the development of cataract in OXYS rats. The data are presented as a percentage of eyes at the appropriate stage of the disease. 0: no change of the lens; 1, 2, and 3: stages of cataract (see Methods). Age of the animals is 12 months; the SkQ1 treatment (one drop of a 250 nM SkQ1 solution daily) was started at age 9 months

Full size image

The third stage of cataract was found in 4 % of eyes in OXYS rats treated with SkQ1 eye drops, already during preliminary examination. The number of the eyes with the third-stage disease was the same upon reexamination 3 months after the start of treatment: we did not expect clarification of lenses with such gross disruptions. At the same time, the number of eyes with second-stage cataract was reduced from 50 to 25 %, whereas with the first stage increased from 46 to 67 %. In 4 % of eye lenses, pathological changes were not detectable. Statistical analysis using the method of dependent pairwise comparison of the cataract stages before and after treatment confirmed that SkQ1 eye drops reduced the stage of cataract (p < 0.001).

ANOVA showed that before treatment, the groups did not differ in the stage of cataract (1.50 ± 0.50 and 1.53 ± 0.57 of AREDS units [mean ± SD]; p = 0.812). Ophthalmoscopic examination after SkQ1 treatment showed that the stage of cataract was reduced compared to the control (1.30 ± 0.59 and 1.68 ± 0.547 of AREDS units [mean ± SD] respectively; p < 0.016). Thus, the SkQ1 eye drops were moderately effective against cataract because Cohen’s d (effect size based on standard deviation) was 0.68.

Effects of SkQ1 treatment: histological and ultrastructural analysis

In the untreated group of OXYS rats, we identified typical (for age 12 months) structural alterations of the lens capsule under a light microscope (Fig. 2): substantial differences in the thickness of capsules (this change is a sign of cataract), and seals and cracks in the capsule body. We can assume that the cause of such changes is a failure to form a 3-dimensional network of protein capsules; this failure can be caused by aberrant posttranslational modifications of proteins and proteoglycans in the lens capsule. The lens epithelial layer of untreated OXYS rats was different from that in untreated Wistar rats: a more jagged edge of the nucleus, and fewer vesicles. Treatment with SkQ1 eye drops attenuated the interstrain differences. The capsules varied in thickness, and pronounced structural changes were not identified. Only occasional mild age-related alterations were detected in the lenses of both control and SkQ1-treated groups. As a result, the condition of lens epithelial cells of SkQ1-treated OXYS rats was approaching the condition of epithelial cells in (untreated) Wistar rats. Fibers in the lenses of untreated OXYS rats had the characteristic mosaic coloration, and changes of the fiber size were observed (increases and decreases) along with various disturbances of the hexagonal structure and packing of the fibers (Fig. 2). The fibers in the lenses of SkQ1-treated OXYS rats had a more correct hexagonal shape in comparison with the untreated OXYS rats.

Fig. 2
figure 2

Comparison of lens morphology of untreated Wistar (a, b, and c) and untreated OXYS rats (d, e, and f) and the lens capsule of OXYS rats treated with SkQ1 eye drops (g, h, and i). The age of all rats is 12 months. Black arrows show disruption of the packing of lens fiber cells, and white arrows indicate disturbances in the lens capsule. The lens capsule of Wistar rats appears homogeneous, and the lens capsule of OXYS rats starts to show areas of “seals” with age; the capsule of the lens treated with SkQ1 appears homogeneous. The fiber cells of Wistar rats have a regular packing geometry. The fiber cells of intact OXYS have typical (for this age and stage of disease) deviations; fiber cells of OXYS rats treated with SkQ1 are arranged into a regular 3-dimensional hexagonal packing pattern. The Venetian blind artifact in d was caused by vibration of a tissue block/knife. Azure II staining; the scale bar is 10 μm

Full size image

Comparative evaluation of ultrastructure of the lens epithelial cells also revealed significant differences between the untreated OXYS and untreated Wistar rats (Fig. 3). The epithelial cells in the germinative region of Wistar rats (Fig. 3a and b) showed granular cytoplasm with short cisterns of the rough endoplasmic reticulum (ER), a small Golgi apparatus, and few lysosomes. Prominent folds of the basolateral plasma membrane and widened intercellular space were frequent in the epithelial layer in untreated Wistar rats, as was endocytosis of the membrane folds. Although nuclear structure was similar in the two rat strains, the cytoplasm in the epithelial cells of untreated OXYS rats looked “empty”: cell organelles were scarce (Fig. 3c and d). Although a cross-section of the cytoplasm of epithelial cells in Wistar rats contained 10–15 ER cisterns, a similar cross-section of the cytoplasm in the cells of untreated OXYS rats contained only 2–3 cisterns. Folds of the basolateral plasma membrane were less developed (one to three folds were found along the plasma membrane from apical to basal surface, in contrast to four to five folds in Wistar rats). Widening of the intercellular space was evidently smaller, and the epithelial layer in untreated OXYS rats looked less “vesiculated.” The reduced folding of the plasma membrane reflects a slowing of transport processes across the membrane. The plasma membrane in the apical zone of untreated OXYS rats frequently had high electron density suggestive of alterations. Many cells contained small osmiophilic bodies in the cytoplasm (Fig. 3d), which represent so-called residual bodies and are related to alterations in the intracellular structure. The ultrastructure of epithelial cells was indicative of a slowing of the metabolism and degenerative changes in the lens of 1-year-old untreated OXYS rats, while the epithelial layer in the lens of Wistar rats showed normal morphology.

Fig. 3
figure 3

Epithelial cells of the lens in untreated Wistar (a, b) and OXYS rats (c, d) and in OXYS rats (e, f) treated with SkQ1 eye drops (250 nM). The age of all rats is 12 months. Transmission electron microscopy of ultrathin sections. Epithelial cells in OXYS rats show the absence of cytoplasmic structures and tight contacts with neighboring cells. 1: a lens capsule adjacent to the basal plasmalemma, 2: cortical lens fibers adjacent to the apical plasmalemma of an epithelial cell, and 3: the nucleus. Asterisks show cisternae of ER; black arrows show the intercellular space among basolateral outgrowths; white arrows: the lateral plasmalemma; arrowheads: myelinlike structures

Full size image

Electron microscopy (lenses of OXYS rats treated with SkQ1 drops) revealed that the cells of the lens germinative region had enlarged intercellular space in the basolateral region (“friable” basal labyrinth); this enlargement became substantial: up to 1.5–2.0 μm in width (Fig. 3e and f). A noticeable increase in endocytosis of plasma membrane folds was not observed. Another striking change in the structure of the epithelial cells of the germinative region was a large amount of rough ER in the cytoplasm (15–20 cisterns could be found in a cross-section of the cytoplasm); the surface of ER was connected with clusters of granular material, as was the case in untreated Wistar rats. The degree of development of rough ER in the epithelial cells of the germinative region was visibly higher compared to the cells of untreated OXYS rats, and approximately the same as in Wistar rats.

In the epithelial cells of the germinative region of SkQ1 eye drop–treated OXYS rats, the number of osmiophilic structures was considerably smaller compared to the untreated OXYS rats. Osmiophilic structures were absent in the sections of the majority of cells, and those that were found had a comparatively small size. Osmiophilic areas of the plasma membrane were absent in the contact zone of the apical plasma membrane of cells of the epithelium and of fiber cells.

Effects of SkQ1 eye drops on α-crystallin gene expression

These effects were evaluated in rats of both strains. Two-way ANOVA showed that αA-crystallin mRNA expression in the lenses was not affected by genotype (strain), but was affected by SkQ1 drops (F1,23 = 8.67, p = 0.008), and those two factors were interacting (F1,23 = 5.84, p = 0.006). Post-hoc analysis revealed a decreasing mRNA level of the αA-crystallin gene in the lens of untreated OXYS rats (F1,12 = 15.01, p = 0.004) compared to the Wistar strain. SkQ1 drops did not change αA-crystallin mRNA expression in Wistar rats, but increased it in OXYS rats (F1,12 = 41.45, p = 0.000) to the normal level of Wistar rats (p = 0.056).

Expression of the αB-crystallin gene was affected by genotype (F1,23 = 230.81, p < 0.0001), and was not influenced by SkQ1 eye drops, but those two factors were interacting (F1,23 = 81.06, p < 0.0001). One-way ANOVA revealed a lower level of αB-crystallin mRNA expression in the lenses of untreated OXYS rats compared to untreated Wistar rats (F1,11 = 191.25, p < 0.0001). Treatment with SkQ1 drops did not affect the αB-crystallin mRNA expression in Wistar rats, but increased it in OXYS rats (F1,11 = 19.33, p = 0.001), albeit not to the normal level of Wistar rats (F1,11 = 165.04, p < 0.000). As a result (Fig. 4), in the lens of OXYS rats treated with SkQ1 drops, the transcriptional activity of the αA-crystallin gene was 2.7-fold higher compared to untreated rats, and reached the level of Wistar rats. αB-Crystallin expression increased 3.4-fold compared to the untreated OXYS rats, but was still 5-fold lower than in untreated Wistar rats (Fig. 4). Moreover, a negative correlation was found between the signs of cataract in OXYS rats and αA-crystallin and αB-crystallin gene expression levels (Spearman’s r = −0.53, p < 0.009, and r = −0.67, p < 0.001, respectively).

Fig. 4
figure 4

Relative gene expression levels of αA- and αB-crystallin genes in lens epithelial cells. The figure shows effects of SkQ1 eye drops from 9 to 12 months of age on mRNA expression of αA- and αB-crystallin in the lens OXYS rats (compared to untreated Wistar rats, according to real-time PCR, n = 6 to 7; the data are normalized to Rpl30 gene expression and presented as mean ± SEM). The level of expression (a relative fluorescence signal in the PCR reactions) is expressed in arbitrary units. Statistical significance (p < 0.05): #interstrain differences, *effects of SkQ1

Full size image

Western blot data on αB-crystallin protein in the lens of OXYS and Wistar rats are shown in Fig. 5. The lens cells of OXYS rats at the age of 12 months showed a significant decrease (by 25.4 %) in the amount of αB-crystallin compared to Wistar rats of the same age (F1,11 = 7.03, p = 0.036; Fig. 5). After treatment with SkQ1 eye drops, there was a significant increase (by 23 %) in the amount of αB-crystallin in the lens cells of OXYS rats (F1,11 = 14.53, p = 0.005).

Fig. 5
figure 5

Alterations in αB-crystallin levels and effects of SkQ1 on OXYS rats. The figure shows Western blot analysis of αB-crystallin in the lens of rats. The relative quantity of the αB-crystallin protein was calculated as intensity of an αB-crystallin band divided by intensity of the corresponding β-actin band. The data are shown as mean ± SEM. Statistical significance (p < 0.05): #interstrain differences, *effects of SkQ1

Full size image

Discussion

The primary purpose of this study was validation of anticataract effects of SkQ1 eye drops on OXYS rats as an animal model of the disease. OXYS rats are already being used to study the mechanisms of cataract formation [13, 20, 21] and to evaluate the efficacy and possible side-effects of new treatments of this disease [7, 22–24] including SkQ1 [8, 25–27]. Previously, to evaluate the therapeutic effects of SkQ1 eye drops and of nutritional supplementation with SkQ1, we used clinical assessment by means of ophthalmoscopy and biomicroscopy only [7–9]. Here, along with the clinical evaluation, histological examination convincingly confirmed that administration of eye drops containing submicromolar concentrations of SkQ1 reverses pathological changes in middle-aged OXYS rats.

Previously, using light and electron microscopy, we demonstrated morphological and pathological changes involving all structural components of the lens in OXYS rats, and we showed that they correspond to senile cataract [28, 29]. In support of our previous reports, we show here that OXYS rats develop cataract resembling human senile cataract with typical disruption or disappearance of normal fiber structure in the opaque region of the lens. All lenses of control OXYS rats exhibit aberrations of varying severity; these problems include large vesicles in cortical fiber cells, swelling and disorganization of fiber cells, and defective migration and elongation of the fiber cells. All of the above leads to worsening of lens transparency. These alterations are thought to be related to dysfunction of transmembrane and cytoskeletal proteins, such as vimentin, nestin, tropomodulin, ephrin, and others [30–32], as well as to disruption of water and ion exchange homeostasis [33, 34]. Furthermore, lens fiber cells in OXYS rats appear rounded and irregular in a cross-section, in contrast to their normal hexagonal appearance in Wistar lenses.

It is known that enhancement of vesicular transport may be associated with disruption of transport systems in the lens, disturbances of the osmotic balance, accumulation of water in the vesicles [35], or with activation of the processes that remove collapsing cells from the cortical layer [36]. The formation of intracellular “vesicles” and expansion of the intercellular space are associated with excessive hydration of the lens, and are typical for lens fibers in various forms of cortical cataract [37].

In the present work, ophthalmoscopic examination shows that the mitochondria-targeted antioxidant SkQ1 completely prevents the cataract development in OXYS rats, and in some of the animals, it reduces signs of the disease in the lens. As expected, we did not see clarification of lenses in rats with third-stage cataract and markedly impaired vision. Such aberrations are detectable in 4 % of eyes in OXYS rats before and after SkQ1 treatment. At the same time, the number of eyes with the second stage of cataract falls from 46 to 25 %. Some mechanisms of this therapeutic action of SkQ1 eye drops are explored in this study.

Treatment with SkQ1 eye drops was started at the age of 9 months, and it elevates the number of “vesicles” in the lenses of both Wistar and OXYS rats at the age of 12 months. These data may be indicative of enhancement of metabolism of damaged macromolecules and cellular components. As a result, SkQ1 significantly improves the size and packing of cortical fibers in OXYS rats, and thus prevents the development of lens opacification in the cortical areas. It is possible that this effect is due to improvement of the state of cells in the epithelial layer.

The lens capsule is a modified basement membrane that surrounds the ocular lens. The capsule of the avascular lens serves as the sole filter for the transit of solutions and molecules between the ocular environment and the lens cells [38, 39]. Abnormal molecular architecture of the lens capsule may cause spatial variation in its permeability to macromolecules. According to the literature, during aging, there are structural changes in lenticular collagen with a concomitant decrease in solubility; this alteration may play some role in deprivation of the lens of its essential nutrients, thereby causing deficiencies that initiate the development of opacity [40]. Structural changes caused by the damage to proteins are compounded by decreased synthesis of extracellular matrix proteins in the lens epithelium. Accumulation of damaged collagen in the capsule itself can lead to activation of a stress signaling pathway called unfolded protein response and to cellular changes underlying cataract formation [41]. At the severe stages of cataract in OXYS rats, we observe characteristic alterations of the lens capsule, which can cause disturbances in the 3-dimensional network formed by a major protein of the lens capsule, collagen IV [41]. As a consequence, a loss of strength, bad permeability, and impaired protective properties of the capsule are expected to ensue. According to the present results, SkQ1 eye drops significantly reduce the anomalies in the lens capsule of OXYS rats. This effect can be a consequence of general improvement of the epithelium functionality and thus of upregulation of extracellular matrix proteins.

In the germinative region of 1-year-old untreated OXYS rats, electron microscopic analysis of the lens epithelial cells reveals weakening of metabolic and biosynthetic processes compared to age-matched untreated Wistar rats. The structure of the cells clearly shows a lowered level of protein synthesis and signs of destructive processes. The “seal” of the basal labyrinth is suggestive of reduced intensity of transport processes. This downregulation of biosynthetic processes can be interpreted as a consequence of ER stress as a result of accumulation of misfolded proteins. Aberrant posttranslational modifications of lens proteins can cause the unfolded protein response and ER stress, which, according to recent data, perform an important function in cataract pathogenesis [41–44]. Overloading of ER with unfolded or misfolded proteins can cause ER stress and activate the unfolded protein response [44, 45].

In the present work, electron microscopic analysis of the cells shows a low degree of development of the rough ER in the OXYS lens compared to Wistar rats; this deficiency may be due to prolonged exposure to ER stress. Our data show that treatment of lens epithelial cells with SkQ1 normalizes protein synthesis and the development of the rough ER, and thereby may contribute to recovery of lens transparency at early stages of cataract development.

An increase in the number of osmiophilic structures in the cytoplasm of lens epithelial cells takes place in the lens of patients with senile cataract; this change may have something to do with the accumulation of damage to cellular structures and with the degenerative processes [46, 47]. The relevant structures are membrane osmiophilic whorls, i.e., degenerative, often autophagic, bodies belonging to the lysosomal class of cytoplasmic organelles, which may occur in a healthy lens but much less frequently than in a cataract-affected one. In the ultrastructural analysis of OXYS lenses, we uncovered an elevated number of osmiophilic structures compared to the untreated Wistar rats. In OXYS rats that receive the SkQ1 eye drops, the number of cells containing osmiophilic structures is significantly lower than in untreated OXYS rats; this phenomenon may reflect a decrease in oxidative damage to cellular components.

During the development of age-related cataract, in humans, over 1,300 genes are downregulated; among them, 241 genes undergo more than a 5-fold diminution of mRNA expression; α-crystallin genes are also usually downregulated [48]. We showed previously that the development of cataract in OXYS rats is linked to substantial downregulation of αA- and αВ-crystallins in the lens [12]. Recently, using high-throughput RNA sequencing (RNA-Seq), we also uncovered extreme downregulation of mRNA of α-crystallin subunits in the OXYS retina during development of AMD-like retinopathy in OXYS rats [49]. Here we confirmed that in the lens of 1-year-old untreated OXYS rats, the levels of mRNA of α-crystallins are significantly lower than in untreated Wistar rats. Treatment with SkQ1 eye drops from the age of 9 months increases the expression of αA-crystallin in the lens of OXYS rats to the level of untreated Wistar rats and significantly increases the level of αB-crystallin, but expression of the latter gene remains below the healthy level of Wistar rats.

Changes in gene expression during cataract development are associated with damage to the lens epithelial-cell layer or its enzymatic system. This pathology is typical for senile cataract [49]. Using light and electron microscopy, we demonstrated here that cataract development in OXYS rats is also associated with dystrophic and atrophic changes in the lens epithelium, and with insufficiency of its drainage and plasticity function. The SkQ1 eye drops improve the morphological and ultrastructural parameters of the lens epithelium and enhance the expression of α-crystallin subunits, i.e., the heat shock proteins facilitating the functioning of the ER in lens epithelial cells.

α-Crystallin is a major lens protein, constituting up to 40 % of total lens proteins. In the lens, α-crystallin’s structural function is to support maintenance of the proper refractive index of the lens. α-Crystallin is the predominant member of the family, and is composed of two types of subunits, A and B, which noncovalently associate to form aggregates with an average molecular weight of 800 kD. α-Crystallin performs a crucial function in preserving lens transparency, not only as a structural protein but also as a lens chaperone, keeping proteins in their native conformation [50]. Factors that suppress the chaperone function of α-crystallin are thought to accelerate aggregation of other crystallins that are undergoing age-related modifications and are losing their native conformation. Consistent with this hypothesis, decreased chaperone activity, increased crystallin aggregation and light scattering, and a loss of lens transparency have indeed all been demonstrated in the aged human lens [51]. One of the proposed mechanisms behind the formation of cataract in humans is aggregation of crystallins that occurs in a highly oxidative environment [52]. Furthermore, cells deficient in αB-crystallin are susceptible to ER stress-induced cell death, whereas αB-crystallin overexpression has protective effects [53]. The pathogenesis of senile cataract is thought to involve accumulation of aberrant posttranslational modifications in proteins, including α-crystallin subunits, with the consequent inhibition of their chaperone activity and formation of insoluble aggregates [54, 55].

There are several reports showing a connection between apoptosis of lens epithelial cells and cataractogenesis [56]. We believe that the effectiveness of SkQ1 in the treatment of cataract is linked to the ability of SkQ1 to reduce mitochondrial reactive oxygen species and, as a consequence, to inhibit mitochondria-mediated apoptosis [26] and to improve expression of α-crystallin subunits, i.e., the heat shock proteins facilitating the workings of ER in lens epithelial cells. The diminished formation of free radicals can hypothetically halt the chain reactions caused by the unfolded protein response, and may reduce ER stress.

In general, our data confirm that treatment with SkQ1 eye drops can not only prevent the development of cataract in senescence-accelerated OXYS rats but also partially reverse signs of early or mild cataract. These data indicate that treatment with SkQ1 may be a promising strategy for cataract prevention in humans.