Stefan Löfgren, MD, PhD Karolinska Institutet St. Erik’s Eye Hospital 11282 Stockholm Sweden Tel: 4686723000 Fax: 4686723330 Email: firstname.lastname@example.org
Purpose: To investigate if the previously shown difference in in vivo-induced ultraviolet B radiation (UVR-B) cataractogenesis between pigmented and albino rats can be seen also with in vitro irradiation. The shielding effect of the iris and UVR absorption in the anterior segment is nullified, and inherent differences in lenticular UVR-B sensitivity between the strains may be revealed.
Methods: Lenses from albino (Fischer-344) and pigmented (Brown-Norway) rats were irradiated in vitro with 1.8 kJ/m2 UVR-B. The lenses were cultured in standard environment in a culture incubator. Cataract was quantified daily by measuring the amount of lens forward light scattering over a period of 1 week. All lenses were photographed during the week.
Results: Two days after exposure, both strains developed significant cataract compared to control lenses, and the light scattering increased exponentially to the last day. From day 4, exposed Fischer lenses scattered more light than Brown-Norway lenses. This difference increased towards the end of the week. The type of cataract (anterior subcapsular, equatorial, and posterior cortical cataract) was similar in both strains. No anterior polar or nuclear cataract was observed.
Conclusions: Lenses from albino Fischer rats are more sensitive to in vitro UVR-B than lenses from pigmented Brown-Norway rats. Ultraviolet B radiation cataract type induced in vitro differs from in vivo cataract in pigmented rats, but not from albino rats. In vitro UVR-B exposure induces more cataract than corresponding lenticular UVR-B in vivo exposures, for both albino and pigmented rat.
Cataract is defined as lens opacities disturbing vision. One cause of cataract is ultraviolet radiation (UVR), with the sun as common source. There is both epidemiological support (Bergmanson & Söderberg 1995; West et al. 1998; Delcourt et al. 2000) and experimental evidence (Pitts et al. 1977; Söderberg 1990) for cataract development after UVR exposure. The cornea serves as a UVR-B (280–315 nm) filter for the lens. Similarly, the lens functions as both a UVR-B and a UVR-A (315–400 nm) filter for the retina. The drawback of this filter function is that the lens itself can be damaged by the UVR and develop cataract.
Current safety limits for avoidance of UVR cataract are mainly based on experiments with pigmented rabbits (Pitts et al. 1977). The most common animals used in UVR cataract research have been rabbits, rats, and mice for in vivo experiments, and rat, bovine, and porcine lenses for in vitro experiments. The tradition of using albino animals in science is also widespread in ophthalmic research but not without drawbacks. The major criticism is that humans are pigmented. More studies on pigmented animals are needed to achieve a better knowledge about UVR cataract.
Albino rats are more sensitive to in vivo UVR exposure than pigmented rats (Löfgren et al. 2010). This difference might be caused by differences in UVR transmission of the cornea, aqueous, or the iris, but different antioxidative defense mechanisms may also exist in the anterior segment or in the lenses per se. In vivo experiments have the advantage of allowing the ocular tissues to interact with each other and the body. However, in order to pin down a single factor, in vitro experiments are necessary. The avascular lens is especially suitable for in vitro experiments because it, even in the eye, can be viewed as an organ culture.
The present experiment was undertaken to identify any inherent differences in UVR sensitivity in lenses from pigmented and albino rats. With in vitro UVR exposure, the role of factors outside the lens is nullified.
Female Fischer-344 albino rats (F344/Ntac@Mol) and Brown-Norway pigmented (BN/Mol) rats were purchased from M&B A/S (Ry, Denmark). Ethical approval was obtained from the Northern Stockholm Animal Experiments Ethics Committee. The treatment of the animals adhered to the ARVO Statement on the Use of Animals in Ophthalmic and Vision Research.
The UVR-B source was a 300 W high-pressure mercury lamp (Oriel Instruments, Stratford, CT, USA) equipped with a 10-cm water filter, double monochromator, and collimating optics. The output had a maximum at 302.6 nm with 4.5 nm full width at half maximum (Fig. 1) as measured with a spectrometer calibrated to a grey-body radiator (PC2000; Ocean-Optics, Duiven, the Netherlands). Irradiance was measured with a thermopile system (Oriel Instruments) calibrated to an American National Standard Institute traceable source by the Swedish National Bureau of Standards.
At day 0, the rats were killed by CO2 asphyxiation. The eyes were enucleated and the lenses were extracted by a posterior approach. The lenses were then immediately immersed in 5 ml balanced salt solution (BSS) in sterile plastic culture Petri dishes. The UVR-B transmittance of the bottom of the plastic culture dish was measured on 10 different dishes with a spectrophotometer in order to calculate the dose of UVR incident to the incubated lens. Immediately after immersion of both lenses from each animal, one lens was exposed at room temperature to 1.8 kJ/m2 UVR-B for 30 min. The lenses were positioned anterior side down, and the radiation was directed from beneath using an aluminium mirror.
Immediately after the UV irradiation, both lenses from each rat were transferred to preheated culture medium in dishes with vented lids. The dishes were prefilled with 4 ml Medium 199, with Earle’s salts without phenol red. The culture medium also contained 25 mm HEPES, 0.1 mg/ml l-glutamine, 0.9 mg/ml sodium bicarbonate, 50 μg/ml streptomycin, 50 U/ml penicillin and 8% calf serum. The incubation environment was set to 37°C and humidified air with 5% CO2. The medium was preincubated in the incubation chamber at least 1 hr prior to use, both at the initial addition of medium and the subsequent media changes. The medium was changed every other day. The lenses were photographed during the course of 1 week of culture, with incident and/or with dark-field illumination.
Quantification of cataract
The degree of cataract was quantified with an objective technique developed by Söderberg et al. (1990). Briefly, a probing white light from a cold light source was directed from underneath the lens in a dark-field illumination setting. The forward scattered light from the lens was collected by the optics of a camera equipped with a photodiode in the film plane. The unit for light scattering is tEDC, transformed Equivalent Diazepam Concentration (Söderberg et al. 1990).
Time outside the incubator was about 30 min at each measuring session.
All chemicals for the culture medium were purchased from Sigma (St. Louis, MO, USA), BSS from Alcon AB (Stockholm, Sweden), filters for sterilization from Millipore AB (Solna, Sweden), and Falcon GripEasy culture dishes from Labora AB (Partille, Sweden). Purified water of biomolecular grade was obtained from a Millipore system.
Six rats were used in a pilot culture trial. Twenty-three Brown-Norway and 22 Fischer rats were used in the main experiment. From each rat, one lens was exposed to UVR, while the other served as a control. Each lens was measured for light scattering daily during 7 days. At each occasion, light scattering was measured in triplicate for each lens.
The significance level was set to 0.05, and the confidence coefficient to 95%. The paired differences in lens light scattering between exposed and control lenses were used as primary data in the linear regression and subsequent two sided t-test for the slope parameter.
The change of medium every other day disturbed the lenses. Within a few minutes after change, the lenses developed a transient anterior haze. The haze disappeared after about 15 min. Because of the development of haze at medium change, the light scattering in the lenses was measured immediately before change of medium. In several dishes, a slight clouding of the medium appeared randomly after about 5 days. The clouding reappeared even after change of medium. This occurred in three UVR-exposed and four control Brown-Norway lenses, and in four UVR-exposed and one control Fischer lenses, and never in both lenses from the same animal. The light scattering values for these lenses stayed within the 95% confidence interval for the mean light scattering for the respective group of lenses without media clouding (data not shown).
The nonexposed lenses were clear in incident illumination (Fig. 2), while, in dark-field illumination, they scattered light mainly from the outer equatorial rim (Fig. 2). The average light scattering in the Brown-Norway lenses increased from 0.07 to 0.10 tEDC over the duration of 1 week, and from 0.06 to 0.08 tEDC for the Fischer lenses. After 4–5 days, some lenses developed a faint anterior haze, only discernible in stereoscopic view. The dark-field photography only resolves a slightly marked equatorial rim (Fig. 2).
One day after exposure, some exposed lenses developed a faint hazy-granular subcapsular cataract, confined to the anterior surface (Fig. 3). On day 2, it had progressed and was evident in all exposed lenses. The granules were best viewed from the posterior aspect. On days 2–3, shallow small equatorial vacuoles and deeper cortical opacities posterior to the equator appeared. On days 4–5, the latter opacities progressed to the posterior suture region, creating a triangular clear zone at the pole (Fig. 3). All three types of cataract progressed resulting in: (i) anterior subcapsular cataract with a floccular appearance; (ii) very large shallow equatorial vacuoles; (iii) shell-like opacification of the whole mid-to-deep posterior cortex, not crossing over the equator (Fig. 3). No nuclear cataract was observed.
After a lag phase of 1 day, the light scattering in the exposed lenses increased steadily with time after exposure (Fig. 4). On day 2, the light scattering was significantly stronger in exposed lenses than in nonexposed lenses in both strains, as indicated by the confidence interval for the mean paired-sample difference between exposed and nonexposed lenses (Fig. 4). Lenses from Fischer rats registered the highest readings, following the trend of increasing difference over the course of the week. At day 7, the Fischer group exhibited significantly more light scattering than the Brown-Norway group. The linear regression and t-test revealed a significantly steeper rate of lens light scattering in the Fischer rats (slope 0.073, n = 22) compared to the Brown-Norway rats (slope 0.055, n = 23), p = 0.04.
The 1.8-kJ/m2 UVR dose incident on the lens surface was chosen to mimic the 5-kJ/m2 corneal dose given in the earlier in vivo UVR experiment (Löfgren et al. 2010). The dose correlates to about 5 kJ/m2 corneal in vivo dose using a transmittance of 40% through cornea and 90% through the aqueous (Maher 1978; Dillon et al. 1999). A 5-kJ/m2 corneal in vivo dose is slightly above the dose shown to induce irreversible cataract in rabbits and rats (Pitts 1970; Michael et al. 1998b).
The transient superficial clouding of lenses at each change of medium indicates different environment in the old and the fresh media. Metabolic or pH differences are probably the cause, while difference in temperature can be outruled. Because nonexposed lenses stayed clear during the incubation period, the transient changes do not seem to permanently alter lens physiology. Same type of transient clouding has also been observed in our research group when fresh mouse, rat, or guinea pig lenses are put in BSS, Ringer’s acetate or fresh medium. Furthermore, it is similar to the transient haze occurring in lenses in animals where the eyelids are kept open (Fraunfelder & Burns 1970; Zhang et al. 2007). The exact cause has not been fully demonstrated.
The clouding of the medium might be a sign of contamination, despite the antibiotics. Because of bulky equipment in different locations (UVR source, light dissemination meter and microscope with camera), the lenses could not be handled in the laminar flow environment at all times outside the incubator. The daily measurements introduced several opportunities for contamination because the culture dish lids had to be removed during scattering measurements and photography. However, there was no difference in frequency in clouding between controls and exposed lenses, nor between the Fischer and Brown-Norway strains. Although bacterial cultures were not taken and definite diagnosis of contamination is unavailable, the proportionately small numbers of randomly affected lenses likely do not affect the results of the study.
The lens light scattering (in tEDC units) as measured in the Petri dish is 12% lower than for the cuvette-based measurements in earlier nonculture experiments. Correcting for this, the light scattering presented here is still 3-to-10-fold higher than a previous in vivo experiment (Löfgren et al. 2010). In vivo, albino rats were more sensitive to UVR than pigmented rats, adhering to the hypothesis that the difference was because of an iris shielding effect (Löfgren et al. 2010). Other factors that differ between in vitro and in vivo exposures are: (i) Concentration of ascorbate in the cornea and the aqueous (Reddy et al. 1998; Ringvold 1998a) absorbing UVR and functioning as an antioxidative agent. The ascorbate concentration in nocturnal animals like rats (Ringvold 1998b), although low, may serve a function. (ii) Some of the UVR-B is absorbed in the cornea and aqueous and re-emitted at longer wavelengths as fluorescence or heat (Ringvold 1998a). This factor is however of a very small magnitude. (iii) The oxygen partial pressure in the lens differs between in vitro and in vivo exposure, introducing different load of reactive oxygen species (Kwan et al. 1972). (iv) Inflammatory processes in the anterior eye segment might induce or aggravate UVR cataract (Löfgren et al. 2010). Given all these factors, with a majority providing protection in the in vivo situation, it is not surprising to find more UVR cataract in the present in vitro experiment compared to the preceding in vivo experiment with a similar lenticular UVR dose (Löfgren et al. 2010).
The trend of increasing difference in lens light scattering between the two strains, supported by the significant difference at day 7, indicated a higher photosensitivity in the Fischer rats. The scarcity of data on photoprotective or sensitizing properties in lenses from pigmented and nonpigmented animals precludes well-founded speculations.
The lens is in a more dehydrated state than any other tissue. The small intercellular spaces and regular ordering of the fibre cells are the basis for lens transparency. If the ion (and water) pumps are not functioning properly, water will accumulate, leading to fluctuations in refractive index, expressed as cataract. Because of the cellular coupling within the lens (Jacob 1999), any disturbance in ion balance spreads through the tissue. The anterior granular-like subcapsular cataract is most likely caused by apoptosis and necrosis of the epithelial cells. Both the time frame and physical appearance correlate well with in vivo studies using approximately the same lenticular UVR doses (Andley et al. 1996; Michael et al. 1998a,b, 2000).
The cortical opacity surrounding, but initially not involving, the posterior suture lines is caused by disturbed ion and water transport mechanisms. Because of the short penetration of UVR in the lens (Maher 1978; Dillon et al. 1999; Löfgren & Söderberg 2001), no UVR reaches the posterior lens region. The fibre cells are however elongated, reaching from the UVR-exposed anterior cortex to the posterior. Anterior UVR damage spreads intracellularly to the posterior fibre cell parts. Ultimately, the pump load will be too strong for the ion pumps, and indirectly the mitochondria, concentrated around the posterior sutures (Gorthy & Anderson 1980). As a result, the posterior suture region becomes opacified. The epithelium might have enough pump capabilities to keep the anterior cortical fibre cell parts transparent.
Equatorial vacuoles were present in both strains, in contrast with the earlier in vivo experiment where only the albino rats developed lens vacuoles. These vacuoles, described in a chloride channel inhibitor model by Young et al. (2000), also appear because of malfunctioning ion transport mechanisms. There are regional differences in transport direction in the lens, with the equator having the highest concentration and activity of ion pumps and gap junctions, resulting in a net outward current (Donaldson et al. 2001). When UVR hits the equatorial region, the ion pump function deteriorates, resulting in an accumulation of extracellular water (Michael et al. 2000).
The results presented here suggest that there is a small but existing difference in inherent UVR sensitivity between pigmented and albino rat lenses. The UVR cataract type induced in vitro differs from in vivo cataract in pigmented rats, but not from that in albino rats. The degree of in vitro-induced cataract in both albino and pigmented rats is higher than in equipotent lenticular UVR-B in vivo exposure, probably because of shielding effects of the lens equator by the iris in vivo.
Many thanks to Professor Per Söderberg for constructive comments and Drs Manoj Kakar and Vino Mody Jr for linguistic support. Financial support was generously provided by Eir 50th Anniversary Foundation, Sigurd and Elsa Golje Memorial Foundation, Anders Otto Swärd Foundation, Swedish Society for Medical Research, Swedish Radiation Protection Institute, and Hildur Pettersson Foundation.