Filtering blue light reduces light-induced oxidative stress, senescence and accumulation of extracellular matrix proteins in human retinal pigment epithelium cells

Authors


  • Conflict/competing interest: No stated conflict of interest.

  • Funding sources: No specific funding.

Dr Marcus Kernt, Department of Ophthalmology, Ludwig Maximilian University, Mathilden St 8, 80336 Munich, Germany. Email: marcus.kernt@med.uni-muenchen.de

Abstract

Background:  Cumulative light exposure is significantly associated with ageing and the progression of age-related macular degeneration. To prevent the retina from blue-light damage in pseudophakia, blue light-absorbing intraocular lenses have been developed. This study compares the possible protective effects of a blue light-absorbing intraocular lens to an untinted ultraviolet-absorbing intraocular lens with regard to light-induced oxidative stress and senescence of human retinal pigment epithelium.

Methods:  As primary human retinal pigment epithelium cells were exposed to white light, either an ultraviolet- and blue light-absorbing intraocular lens or ultraviolet-absorbing intraocular lens was placed in the light beam. After 60 min of irradiation, cells were investigated by electron microscopy for viability, induction of intracellular reactive oxygen species, and senescence-associated β-galactosidase activity. Expression and secretion of matrix metalloproteinases 1 and 3 and their mRNA were determined by real-time polymerase chain reaction (PCR) and enzyme-linked immunosorbent assay.

Results:  Light exposure induced structural damage, decreased retinal pigment epithelium cell viability, and increased reactive oxygen species, senescence-associated β-galactosidase activity and matrix metalloproteinases 1 and 3 expression and secretion. Although both types of intraocular lens significantly reduced these effects, the protective effects of the ultraviolet- and blue light-absorbing intraocular lens were significantly stronger than those of the ultraviolet-absorbing intraocular lens.

Conclusions:  The ultraviolet- and blue light-absorbing intraocular lens demonstrated significantly better protection against light-induced oxidative stress, senescence and structural damage than the ultraviolet-absorbing intraocular lens. These in vitro findings support the hypothesis that the ultraviolet- and blue light-absorbing intraocular lens may prevent retinal damage in clinical use.

Introduction

Age-related macular degeneration (AMD) is the major cause for blindness in people over 60 years of age and is a socioeconomic burden in developed countries.1 Both dry and wet forms of AMD exist. Wet AMD can result in rapid loss of central vision due to choroidal neovascularization in the central retina.2 However, only about 10% of AMD patients suffer from wet AMD.2 The other 90% of patients suffer from dry forms, such as geographic atrophy,2 in which visual acuity usually decreases gradually. Nevertheless, the gradual decrease of central vision often impairs the ability to read or drive a car. At present, there is no satisfactory treatment for geographic atrophy.1

Geographic atrophy is characterized by the atrophy of retinal pigment epithelium (RPE) cells and consecutive photoreceptor degeneration.1 Although the exact pathogenesis remains unclear, AMD is a disease of the elderly population and the senescent degeneration of RPE cells due to ageing processes is likely a major reason for the development of geographic atrophy.

A common characteristic of senescent cells is the accumulation of extracellular matrix (ECM), which can be a result of remodelling during the ageing process.3 When such changes occur, several consequences of the disorganization or functional alteration of connective tissue are likely. For example, ageing is associated with progressive thickening of the RPE and Bruch's membrane due to deposition of matrix components and membranous debris rich in lipids.4 A consequence of these ageing processes is an exponential decline in the hydraulic conductivity, which may lead to malnutrition and degradation of cells.3

Even if epidemiological studies are inconclusive,5–7 there is evidence that cumulative light-exposure is significantly associated with progression of AMD.6,8,9 In addition, cumulative light exposure is significantly associated with the premature ageing of cells.3

Both human and animal studies suggest that it is mainly the short wavelengths within the ultraviolet (UV) and blue light ranges that result in retinal damage and can induce photochemical damage and oxidative stress to the RPE.10,11 The accumulation of oxidative damage is a major risk factor for premature ageing of cells.3 In addition, epidemiological studies seem to indicate that aphakic and pseudophakic eyes have an increased risk for the development and progression of AMD.8,12,13 Light-induced ageing processes within the RPE might be responsible for this.

This study compares the possible protective effects of UV- and blue light-absorbing intraocular lenses (IOLs) and untinted UV-absorbing IOLs with regard to light-induced oxidative stress and senescence in human RPE cells.

Methods

Ethics

The methods of securing human tissue complied with the Declaration of Helsinki and were approved by the local ethics committee. Written informed consent was obtained from the tissue donors.

Human RPE

The RPE cells from five human donors (44, 48, 58, 68 and 75 years old, collected 3–10 h post-mortem) were obtained from the eye bank at Ludwig Maximilian University and prepared as previously described.14 Briefly, whole eyes were thoroughly cleansed in 0.9% NaCl solution, immersed in 5% polyvinyl pyrolidone iodine, and rinsed again in 0.9% NaCl solution. The anterior segment of each donor eye was removed, and the posterior poles were examined with the aid of a binocular stereomicroscope to confirm the absence of gross retinal disease. The neural retina was then carefully peeled away from the RPE-choroid-sclera complex with fine forceps. The eye cup was rinsed with Ca2+- and Mg2+-free Hanks balanced salt solution and filled with 0.25% trypsin (Invitrogene, Carlsbad, CA, USA) for 30 min at 37°C. The trypsin was carefully aspirated and replaced with Dulbecco modified Eagle medium (DMEM; Biochrom, Berlin, Germany) supplemented with 20% foetal calf serum (Biochrom). The medium was gently agitated with a pipette, releasing the RPE cells into the medium without damaging Bruch's membrane. The RPE cell solution was transferred to a 50-mL flask (BD, Franklin Lakes, NJ, USA) containing 20 mL of DMEM supplemented with 20% foetal calf serum and maintained at 37°C and 5% carbon dioxide. Epithelial origin was confirmed by immunohistochemical staining for cytokeratin with a pan-cytokeratin antibody (Sigma-Aldrich, Steinheim, Germany).14 The cells were tested and found to be free of contaminating macrophages (anti-CD11; Sigma-Aldrich) and endothelial cells (anti-von Willebrand factor; Sigma-Aldrich) (data not shown).

After growing to confluence (100%), primary RPE cells were subcultured and maintained in DMEM supplemented with 10% foetal calf serum at 37°C and 5% carbon dioxide. For all experiments, early passages of primary human RPE cells (passages 2 and 3) were used. All investigated RPE cell cultures still presented intracellular pigmented granules under a phase-contrast microscope. In addition to melanin and other pigments, these granules contain lipofuscin, which is lost from RPE cells cultured long term.14 Lipofuscin is known to act as a cell photosensitizer and to play roles in atrophic degeneration of the RPE and the development and progression of several degenerative retinal diseases.1 For all cell culture experiments, RPE cells were seeded in 35-mm tissue culture dishes and cultured upon confluence in darkness. Before illumination, RPE cells were kept for 24 h in serum-free conditions.

IOLs

Two single-piece acrylic foldable IOLs were tested: (i) a yellow-tinted, UV- and blue light-absorbing IOL (SN60AT, +20 dpt, Alcon Laboratories, Inc., Fort Worth, TX, USA); and (ii) an untinted UV-absorbing IOL (SA60AT, +20 dpt, Alcon Laboratories).

Illumination of cells

A spot-light source (LC-8, Hamamatsu Photonics, Hamamatsu, Japan) from a mercury-xenon lamp using an optic fibre as the light guide (spectral range: 400–700 nm) was used for illumination. The cell culture medium was replaced with phosphate-buffered saline (PBS) solution just before illumination. The plastic cover of the culture well was removed, and the cells were illuminated from above for 60 min (350 mW/cm2) in the presence or absence of the IOLs. This corresponds to approximately the threefold to fourfold intensity of sunlight at noon time in summer in central Europe. Each IOL was applied to the light-emitting output of the optic fibre (Ø 3.5 mm), where it remained attached without an aid. The illumination power and spectral range was measured with a spectrometer (C10083MD, Hamamatsu Photonics). Directly after illumination, the PBS was replaced with serum-free cell culture medium and the cells were kept in darkness for 24 h until analyses were performed. During and after illumination, no significant changes (≤0.5°C) in temperature cell culture medium/PBS could be detected.

Investigating ultrastructural changes by transmission and scanning electron microscopy

For transmission and scanning electron microscopy, RPE cells were cultured in 35-mm tissue culture dishes and treated as describe d above. Cells were postfixated in osmium tetroxide 2% (Dalton's fixative), dehydrated in graded concentrations of ethanol, and embedded in Epon 812. Preparation for light microscopy followed with semithin sectioning of 400 nm and staining with an aqueous mixture of 1% toluidin blue and 2% sodium borax. For transmission electron microscopy, ultrathin sections of 70 nm were obtained by series sectioning and were contrasted with uranyl acetate and lead citrate. Analysis and imaging of five grids per specimen with six to nine sections per grid (30–45 sections per specimen) was performed using a light microscope (Leica DM 2500; Leitz, Wetzlar, Germany) and an electron microscope (EM 9 S-2; Carl Zeiss, Oberkochen, Germany).

For scanning electron microscopy, RPE cells were stained with uranyl acetate and lead citrate and observed using a scanning electron microscope (JSM-6300 I, Jeol, Japan). For scanning electron microscopy, from each treatment group, at least five scans from three donors were graded.

Detection of intracellular reactive oxygen species (ROS)

Intracellular ROS production was detected as previously described.15,16 RPE cells were cultured in 35-mm tissue culture dishes and treated as described above. After treatment, the cells were loaded with 10 µmol/L 5-(and-6)-chloromethyl-2′,7′-dichlorodihydro-fluorescein diacetate acetyl ester (CM-H2DCFDA; Molecular Probes, Eugene, OR, USA) dissolved in Krebs–Ringer bicarbonate buffer (135 mmol/L NaCl, 3.6 mmol/L KCl, 10 mmol/L HEPES, 5 mmol/L NaHCO3, 0.5 mmol/L NaH2PO4, 0.5 mmol/L MgCl2, pH 7.4) for 30 min, incubated with DMEM medium for 2 h, and analysed with an epifluorescence microscope (Aristoplan, Zeiss, Oberkochen, Germany)

MTT assay

The tetrazolium dye-reduction assay (MTT, 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide) was used to determine metabolic activity of RPE cells after illumination. The MTT assay, which is well established for an indirect assessment of cell viability, was performed as described by Mosmann, with some modifications.14,17 The medium was removed, cells were washed with PBS, and 1000 µL of MTT solution (1.5 mL MTT stock, 2 mg/mL in PBS, plus 28.5 mL DMEM) was added to each well. RPE cells were incubated at 37°C for 1 h. The formazan crystals that formed were dissolved by the addition of 1000 µL of dimethyl sulphoxide per well. Absorption was measured by a scanning multiwell spectrophotometer at 550 nm (Molecular Probes). The results are expressed as the mean percentage of proliferation as compared with the control. The control cells were RPE cells of the same passage, which were kept in darkness, without exposure to any illumination. The experiments were performed in triplicate and repeated three times.

Detection of senescence-associated β-galactosidase (SA β-Gal) activity

The proportion of RPE cells positive for SA β-Gal activity was determined as described by Dimri et al.18 Briefly, treated RPE cells were washed twice with PBS and fixed with 2% formaldehyde and 0.2% glutaraldehyde in PBS (pH 6.0) at room temperature for 4 min. The cells were then washed twice with PBS and incubated under light protection for 8 h at 37°C with fresh SA β-Gal staining solution (1 mg/mL 5-bromo-4-chloro-3-indoyl-β-D-galactopyranoside, 40 mmol/L citric acid–sodium phosphate solution [pH 6.0], 5 mmol/L potassium ferrocyanide, 5 mmol/L potassium ferricyanide, 150 mmol/L NaCl, and 2 mmol/L MgCl2 diluted in PBS). Cells were then examined for the development of blue colour and photographed at low magnification (×100) using a light microscope. The results are expressed as mean ± standard deviation (SD) of 12 experiments using RPE cultures from four donors.

Detection of MMP (Metalloproteinase)-1 and MMP-3 secretion by RPE cells

Levels of MMP-1 and activated MMP-3 in the RPE cell culture supernatants were determined by enzyme-linked immunosorbent assay (ELISA). The supernatants were collected 24 h after the illumination treatment. To detect MMP-1 secretion a Quantikine ELISA Assay Kit (R&D Systems, Minneapolis, MN, USA) was used and to detect activated MMP-3 secretion a MMP-3 Biotrak ELISA Assay Kit (GE Healthcare, Buckinghamshire, UK) was used according to the instructions of the manufacturers. Optical density of each 96-well plate was read at 405 nm on a microplate reader (Versamax, Molecular Devices, Sunnyvale, CA, USA). The experiments were performed in duplicate and repeated four times.

RNA isolation and real-time (RT)-PCR

Total RNA was isolated from RPE cells cultured in 10-cm Petri dishes by the guanidium thiocyanate–phenol–chloroform extraction method (Stratagene, Heidelberg, Germany). The structural integrity of the RNA samples was confirmed by electrophoresis in 1% Tris–acetate–EDTA (Ethylenediaminetetraacetic acid) agarose gels. The yield and purity were determined photometrically (BioPhotometer; Eppendorf, Hamburg, Germany). mRNA was isolated and then converted to cDNA via reverse transcriptase. This cDNA was then used for RT-PCR. The quantification of MMP-1 and MMP-3 mRNA was performed with specific primers (Table 1) with a LightCycler instrument (Roche Diagnostics, Mannheim, Germany). Primers and probes were found with ProbeFinder 2.04 (Roche, http://www.universalprobelibrary.com), and all primers and probes were designed to cross intron–exon boundaries in order to avoid amplification of genomic DNA. All PCR products were sequenced to ensure product validity. Each 14-µL reaction volume contained 1 × FastStart DNA Master Hybridization Probes Mix (Roche Diagnostics), 4 mmol/L MgCl2, 0.5 mmol/L of each primer, 0.2 mmol/L TaqMan probe and 2 µL cDNA.

Table 1.  Primers used for RT-PCR
Target GenePrimerPositionSequence
  1. MMP, Metalloproteinase; RT-PCR, Polymerase Chain Reaction.

MMP-1Left1079–11005′-CAGAGATGAAGTCCGGTTTTTC-3′
Right1135–11545′-GGGGTATCCGTGTAGCACAT-3′
MMP-3Left1194–12205′-CAAAACATATTTCTTTGTAGAGGACAA-3′
Right1264–12845′-TTCAGCTATTTGCTTGGGAA-3′

The cDNA of RPE cells exposed to white light in the absence or presence of each IOL was amplified with specific primers for a maximum 40 cycles. The amplification signals were detected in real time, which enabled accurate quantification of the amounts of the initial RNA template. Two oligonucleotides with differently labelled fluorophores were hybridized to the amplified fragment during the annealing phase. When the two probes came in close proximity, fluorescence resonance energy transfer developed between the two fluorophores. The emitted fluorescence was then measured by the LightCycler instrument. Hybridization probes were displaced during the extension step. Depending on the initial concentration of target genes, the signal intensity increased in different cycles, and these cycles were used as the crossing point. The standard curve was made with three different probes of untreated RPE cells. For MMP-1 mRNA signals were detected at a mean of 24 cycles (±3 cycles), for MMP-3 mRNA signals were detected at a mean of 21 cycles (±4 cycles). To normalize for differences in the amount of total RNA added to each reaction, 18S rRNA was simultaneously processed in the same sample as an internal control. The level of MMP-1 and MMP-3 mRNA was determined as the relative ratio (expressed in decimal format), which was calculated by dividing the level of MMP-1 and MMP-3 mRNA by the level of the 18S rRNA housekeeping gene in the same samples. All experiments were performed at least in triplicate and repeated three times.

Statistical analyses

All data were analysed with SPSS 13.0 for Windows (SPSS, Chicago, IL, USA). For all statistical tests, P < 0.05 was considered significant. Results of the MTT assay are presented as mean ± SD units of absorbance. Ten individual samples per group were measured in triplicate, and the Mann–Whitney test was used for analysis of the results. Results from MMP-1 and activated MMP-3 ELISA are presented as mean ± SD ratios of MMP-1 or MMP-3 of each probe, which were normalized to the mean amount of MMP-1 or MMP-3 detected in the control. Results of the RT-PCR are presented as mean ± SD ratios of the investigated mRNA and 18S rRNA. Again, Mann–Whitney testing was applied. All experiments were performed in triplicate and repeated three times.

Results

Ultrastructural findings

According to scanning electron microscopy, after 60 min of illumination with plain white light the number of RPE cells was significantly reduced. In addition, certain cell alterations could be detected. The surface of RPE cells appeared much smoother and had fewer protuberances than the unilluminated control cells. In illuminated RPE cells, the typical cell surface structure with its numerous microvilli was not evident. Instead numerous RPE cells showed constrictions with large open pores, which are typical of apoptotic cells (Fig. 1a).

Figure 1.

Scanning electron microscopy (SEM; a) and transmission electron microscopy (TEM; b) of primary retinal pigment epithelium (RPE) cells with no illumination (a: A, E; b: A), after 60 min of light exposure in the presence of the ultraviolet (UV)- and blue light-absorbing intraocular lens (IOL) (a: B, F; b: B), in the presence of the UV-absorbing IOL (a: C, G; b: C), and after exposure to unfiltered white light (a: D, H; b: D). (a) In RPE cells, the typical cell surface structure with its numerous microvilli was not evident after 60 min of unfiltered illumination. Instead numerous RPE cells showed constrictions with large open pores, which are typical of apoptotic cells were seen (a: D). These effects were significantly reduced by both tested IOLs. However, when light was filtered by the UV- and blue light-absorbing IOL, the effects were significantly stronger reduced than when filtered with an UV-absorbing IOL only (a: B, C). (b) When RPE cells were illuminated with unfiltered light, the cytoplasm and intracellular structures appeared more loosened and dispersed. Intracellular components and cell organelles, such as mitochondria and ribosomes, were swollen or seemed to have burst (b: D). Both tested IOLs were able to reduce these effects, but in the presence of the UV- and blue light-absorbing IOL, the alterations were significantly less pronounced than in the presence of the UV-absorbing IOL (a: B and C). Original magnification for SEM: A–D ×500 and E–H ×2000; TEM: A–D ×9500.

According to transmission electron microscopy, the cytoplasm and intracellular structures appeared more loosened and dispersed after 60 min of unfiltered illumination. Intracellular components and cell organelles, such as mitochondria and ribosomes, were swollen or seemed to have burst (Fig. 1b, arrows).

These light-induced alterations to RPE cells were seen to a much lesser extent when light was filtered with one of the tested IOLs. In the presence of the UV- and blue light-absorbing IOL, the alterations were significantly less pronounced than in the presence of the UV-absorbing IOL (Fig. 1a,b).

Viability of cells

When cells were illuminated with unfiltered white light (350 mW/cm2) for 60 min, a significant reduction of RPE cell viability was detected (34 ± 4.6% viable cells; Fig. 2). When a UV-absorbing IOL was placed in the light beam, this decrease in viability was less pronounced (54 ± 5.8% viable cells; Fig. 2) and significantly different as compared with cells illuminated with unfiltered white light. In the presence of the UV- and blue light-absorbing IOL, this viability decrease was significantly less (76 ± 3.3% viable cells; Fig. 2) than in cells illuminated with unfiltered white light and those illuminated in the presence of the UV-absorbing IOL (Fig. 2).

Figure 2.

Viability of retinal pigment epithelium (RPE) cells after 60 min of illumination with white light (spectral range: 400–700 nm) and in presence of the ultraviolet- and blue light-absorbing intraocular lens (YL) or the ultraviolet-absorbing intraocular lens (WL), as measured by a colorimetric test (MTT) (*P < 0.05; error bars represent SD).

Intracellular ROS

In untreated RPE cells only a faint staining for intracellular ROS was detected, whereas after 60 min of white light exposure intracellular ROS staining was markedly increased (Fig. 3a). When a UV-absorbing IOL was placed in the light beam during illumination, staining for intracellular ROS was noted, but it was markedly less pronounced compared with cells that were illuminated with unfiltered white light. In the presence of the UV- and blue light-absorbing IOL, this reduction of ROS staining was even more distinct.

Figure 3.

(a) Detection of intracellular reactive oxygen species (ROS) (top) and staining for β-galactosidase activity (bottom) after 60 min of illumination with white light (spectral range: 400–700 nm) and in presence of the ultraviolet- and blue light-absorbing intraocular lens (YL) or the ultraviolet-absorbing intraocular lens (WL). (b) Differences in the β-galactosidase-positive cells among treatments (*P < 0.05). Error bars represent SD.

SA β-Gal activity

In untreated controls, very few RPE cells showed the characteristic blue staining for SA β-Gal (8.4 ± 4.1%; Figure 3a,b). When cells were illuminated for 60 min, a significant increase in SA β-Gal staining was detected (37.6 ± 9.7%). Significantly fewer SA β-Gal-positive cells were noted when cells were illuminated in the presence of the UV-absorbing IOL (26.8 ± 3.2%), and an even greater reduction was noted when the UV- and blue light-absorbing IOL was placed into the light beam (13.4 ± 5.8%).

Expression of MMP-1 and MMP-3 mRNA

The expression of both MMP-1 and MMP-3 mRNA was detected in every sample. The detected mRNA levels were normalized to those of 18S rRNA and are expressed as the relative ratio of MMP-1/18S or MMP-3/18S.

Illumination with white light for 60 min led to a significant increase of MMP-1 and MMP-3 mRNA expression in RPE cells as compared with the untreated control cells (MMP-1: 11.7 ± 1.0-fold; MMP-3: 5.5 ± 1.0-fold; Fig. 4). In comparison to unfiltered illumination, filtering light with either of the tested IOLs significantly reduced this light-induced increase in MMP-1 and MMP-3 mRNA. The level of MMP-1 and MMP-3 mRNA was significantly lower when light was filtered by the UV- and blue light-absorbing IOL (MMP-1: 3.3 ± 0.6-fold, MMP-3: 2.4 ± 0.2-fold) as compared with the UV-absorbing IOL (MMP-1: 6.0 ± 0.6-fold, MMP-3: 3.3 ± 0.3-fold).

Figure 4.

MMP-1 (a) and MMP-3 (b) mRNA expression of retinal pigment epithelium (RPE) cells after 60 min of illumination with white light (spectral range: 400–700 nm) and in presence of the ultraviolet- and blue light-absorbing intraocular lens (YL) or the ultraviolet-absorbing intraocular lens (WL), as measured by quantitative RT-PCR (polymerase chain reaction). The relative ratio of MMP (Metalloproteinase)-1 or MMP-3 mRNA was normalized to 18S rRNA and is expressed in decimal format (*P < 0.05; error bars represent SD).

MMP-1 and activated MMP-3 secretion

Light exposure for 60 min led to a significant increase of MMP-1 and activated MMP-3 secretion in cultured RPE cells (MMP-1: 13.9 ± 2.4-fold, and MMP-3: 4.1 ± 0.1-fold; Fig. 5) as compared with the unilluminated control.

Figure 5.

ELISA results for the light-induced secretion of MMP (Metalloproteinase)-1 (a) and activated MMP-3 (b). Untreated primary human retinal pigment epithelium (RPE) cells (Co), after 60 min of illumination with white light (spectral range: 400–700 nm), and in presence of the ultraviolet- and blue light-absorbing intraocular lens (YL) or the UV-absorbing intraocular lens (WL) (*P < 0.05; error bars represent SD).

The secretion of MMP-1 was significantly lower when cells were illuminated in the presence either of the IOLs. MMP-1 secretion was significantly lower in the presence of the UV- and blue light-absorbing IOL (5.6 ± 1.2-fold) as compared with the UV-absorbing IOL (10.2 ± 1.3-fold). For activated MMP-3 secretion, however, the two IOLs did not have a comparable effect. As compared with cells illuminated with white light, MMP-3 secretion was significantly lower in the presence of the UV- and blue light-absorbing IOL (2.5 ± 0.2-fold), whereas no significant reduction was noted in the presence of the UV-absorbing IOL (3.9 ± 0.4-fold).

Discussion

In any living cell, ageing is an inevitable and complex process involving changes in multiple parameters. AMD is a disease of the elderly population that is closely associated with ageing.1 Loss and degeneration of RPE cells, particularly in the macular centre, is pathognomonic for AMD.1 With advancing age, RPE cells undergo an increase in pleomorphism and accumulate metabolic debris from remnants of the incomplete degradation of phagocytized rod and cone membranes.1 This incomplete degradation of phagocytized photoreceptor membranes results in a continuous increase in intracellular lipofuscin in the RPE cells over time, which causes deterioration in cellular function and makes the retina more sensitive to radiation damage.1,2

The natural lens provides protection against the phototoxic effects of UV and blue light due to its filtering characteristics.19 After cataract surgery, however, this filtering function is lost and additional protection against these hazards is needed. For this purpose yellow-tinted, UV- and blue light-absorbing IOLs have been developed and have been in clinical use for several years. Although their use is controversial, implantation of theses IOLs does not seem to have harmful side effects for patients.20 However, clinical evidence regarding the effectiveness of these IOLs is still very limited.20 To test the protective effects of UV-filtering and blue light-filtering IOLs, several experimental studies examined immortalized RPE cell lines artificially laden with lipofuscin to investigate lipofuscin-triggered phototoxic reactions in the RPE. They found that filtering blue light can significantly attenuate light induced increase of intracellular reactive ROS, growth factor expression and apoptosis.21–25 In our own work, we demonstrated that a blue light-filtering IOL significantly reduce light-induced vascular endothelial growth factor A (VEGF) overexpression and stabilize light-induced increase of BAX/Bcl-2 ratio in primary human RPE cells.14 It has been demonstrated that especially the shorter wavelength within the blue spectrum seem to be most harmful regarding RPE damage.26 All these results support the theory that UV-filtering and blue light-filtering IOLs may be more protective against photochemical damage.

In the present study we investigated the potential protective effects of a UV- and blue light-absorbing IOL compared with an untinted UV-absorbing IOL with regard to structural damage, light-induced oxidative stress and senescence in primary human RPE cells.

Our electron microscopy results show clearly that unfiltered white light induces histomorphological alterations in RPE cells. For example, the surface of the RPE cells was less structured and had significantly fewer microvilli, numerous cells showed characteristics of undergoing apoptosis, intracellular components and cell organelles were swollen, and cytoplasm was dispersed. These light-induced changes were significantly reduced in the presence of either IOL, but changes appeared to be even less pronounced when light was filtered by the UV- and blue light-absorbing IOL.

Increased light exposure induces the generation of ROS and apoptosis in human RPE cells.16 Oxidative stress and the associated production of ROS have been implicated as risk factors for atrophic degeneration of the RPE1 and as major factors in triggering cellular senescence, particularly stress-induced premature senescence.27,28 Oxidative damage resulting from increased intracellular ROS accumulation compromises cells and is revealed in the form of mitotic figures and hypopigmentation of the RPE, which resemble the atrophic changes seen in AMD.1,16 Oxygen radicals are known to decrease the viability of retinal cells.1,3

In accordance with previous investigations,14,21,24 the results of our study clearly demonstrate that light exposure strongly induces intracellular ROS in primary human RPE cells. This increase in oxidative stress was accompanied by a decrease of cellular viability. Both of these effects were significantly reduced by the two types of IOL, although the reduction was significantly greater in the presence of the UV- and blue light-absorbing IOL than the UV-absorbing IOL.

β-Galactosidase is an enzyme produced within cells that accumulates in ageing cells over time.29 Therefore, it has become a valuable, selective and reliable marker for senescent cells, both in vitro and in vivo.16,29,30 This study showed that exposure to white light for 60 min increases SA β-Gal activity in human RPE cells. However, filtering light with either a UV- and blue light-absorbing IOL or a UV-absorbing IOL resulted in a significant reduction of light-induced SA β-Gal activity. This finding suggests that filtering light with a suitable IOL would reduce the risk of premature senescence of the RPE after acute light exposure.

Another common finding in senescent cells is the accumulation of ECM, which is a result of remodelling during the ageing process.3 These changes are associated with a disorganization or functional alteration of connective tissue. MMPs are a family of at least 23 enzymes that are secreted by cells.31,32 They cleave almost all protein components of the ECM and are involved in ECM remodelling.31,32 MMP-1 and MMP-3, with their inhibitors TIMP-1 and TIMP-2, are secreted by RPE cells and serve as indicators of ageing processes.31,32 MMP-1 is the most ubiquitously expressed subtype, and it has been implicated in cancer metastasis and premature ageing of the skin following infrared and UV radiation.33,34 MMP-3 can activate latent MMP-1 and degrade several collagen types, as well as non-collagen matrix proteins, most of which are found in the ECM surrounding the RPE and Bruch's membrane. Increased levels of activated MMP-3 are related to destructive inflammatory diseases, such as juvenile rheumatoid arthritis, as well as proliferative vitreoretinal disorders and AMD.34–36

Our experiments demonstrate clearly that light exposure leads to a significant and strong induction of MMP-1 and activated MMP-3 in primary human RPE cells. In addition, this overexpression of ECM proteins was significantly reduced by both tested IOLs, with the effect of the UV- and blue light-absorbing IOL being significantly greater than that of the UV-absorbing IOL.

In vitro is not in vivo, and it is very clear that these experimental data cannot be transferred one to one on the in vivo situation. In our experimental set up, a single-cell line was exposed to quite an intense amount of irradiation over a short time. Nevertheless, the experimental results of the present study suggest that intense light exposure has the potential to induce premature senescence of RPE cells. In addition, these data indicate that filtering blue light can help to reduce these effects at least under experimental conditions. Even if the direct transferability of these in vitro results remains questionable they may provide another hint of support to the idea that UV- and blue light-absorbing IOLs might be useful in preventing irradiation damage in patients with an increased risk for progression of AMD.

Acknowledgements

The authors thank Katja Obholzer and Renate Scheler, from Department of Ophthalmology, Ludwig-Maximilians-University, Munich, Germany, for excellent technical assistance.

Ancillary