This invited paper is part of the Symposium in Print “Retinal Photodamage.”
Lipofuscin and A2E Accumulate with Age in the Retinal Pigment Epithelium of Nrl−/− Mice†
Article first published online: 28 MAR 2012
© 2012 Wiley Periodicals, Inc. Photochemistry and Photobiology © 2012 The American Society of Photobiology
Photochemistry and Photobiology
Volume 88, Issue 6, pages 1373–1377, November/December 2012
How to Cite
Boyer, N. P., Tang, P. H., Higbee, D., Ablonczy, Z., Crouch, R. K. and Koutalos, Y. (2012), Lipofuscin and A2E Accumulate with Age in the Retinal Pigment Epithelium of Nrl−/− Mice. Photochemistry and Photobiology, 88: 1373–1377. doi: 10.1111/j.1751-1097.2012.01127.x
- Issue published online: 29 OCT 2012
- Article first published online: 28 MAR 2012
- Accepted manuscript online: 15 MAR 2012 12:23AM EST
- Received 18 January 2012, accepted 16 February 2012
Lipofuscin is a fluorescent material with significant phototoxic potential that accumulates with age in the retinal pigment epithelium (RPE) of the eye. It is thought to be a factor in retinal degeneration diseases. The most extensively characterized lipofuscin component, N-retinylidene-N-retinylethanolamine (A2E), has been proposed to be a byproduct of reactions involving the visual pigment chromophore. To examine the impact of the visual pigment and photoreceptor cell type on lipofuscin accumulation, we analyzed the RPE from Nrl−/− mice of various ages for lipofuscin fluorescence and A2E levels. The photoreceptor cells of the Nrl−/− retina contain only cone-like pigments, and produce cone-like responses to photostimulation. The cone-like nature of these cells was confirmed by the presence of RPE65. Lipofuscin was measured with fluorescence imaging, whereas A2E was quantified by UV/VIS absorbance spectroscopy coupled to HPLC. The identity of A2E was corroborated with tandem mass spectrometry. Lipofuscin and A2E accumulated with age, albeit to lower levels compared with wild type mice. The emission spectra of RPE lipofuscin granules from Nrl−/− mice were similar to those from wild type mice, with λmaxca 610 nm. These results demonstrate that cone visual pigments can contribute to the production of lipofuscin and A2E.
Lipofuscin has been shown to accumulate with age in the retinal pigment epithelium (RPE) of the eye (1–4) and other postmitotic cells (5). It is a complex mixture of partially digested lipid and protein components, whose precise composition remains unknown. The material is highly fluorescent. Bis-retinoids constitute a major component (6), the most prominent of which is N-retinylidene-N-retinylethanolamine (A2E; 7,8). Only 2% of lipofuscin by weight can be attributed to protein (9).
Accumulation of lipofuscin requires normal operation of the visual cycle to generate 11-cis retinal (10), the chromophore of the photoreceptor visual pigment (11). A2E and other bis-retinoids have been shown to form from all-trans retinal (6,12,13), the form of the chromophore that is released from photoactivated visual pigment. Such reactions take place in the photoreceptor outer segments, which are phagocytized by the RPE on a daily basis. The phagocytized outer segments are digested by the RPE lysosomes, resulting in the formation and gradual accumulation of lipofuscin (14,15). This phenomenon has been examined mostly in animals with rod-dominant retinas, whereas lipofuscin formation has been studied with biochemical preparations from the outer segments of rod photoreceptors, the cells responsible for dim light vision.
Lipofuscin can mediate the generation of singlet oxygen (16) and light-induced damage (17,18), suggesting that its components have significant photosensitizing potential. The components responsible for the bulk of lipofuscin phototoxicity do not appear to include A2E, whose photosensitizing potential is rather modest (19). Lipofuscin phototoxicity is proposed to underlie blue-light damage and play a role in degenerative diseases of the retina, such as age-related macular degeneration (20,21). The overwhelming majority of the photoreceptor cells of the human retina are rods; however, cones, the cells responsible for bright light and color vision, are concentrated in the macula (22). In humans, the highest levels of lipofuscin are found in the macula (2,4).
To understand the contribution of cone photoreceptors and visual pigments to lipofuscin accumulation, we have used mice that lack the gene encoding for the neural retinal leucine zipper (Nrl), a transcription factor necessary for rod photoreceptor development (23). The photoreceptor cells of Nrl−/− mice contain only cone visual pigments, have a large number of disks open to the extracellular space, and have cone-like responses when stimulated by light (24,25). We found that lipofuscin and A2E accumulate with age in the RPE of Nrl−/− mice, and the fluorescence emission spectrum of Nrl−/− lipofuscin is similar to that of wild type mice. These data imply that indeed cones are capable of producing A2E and lipofuscin, and this may be the source of the high-lipofuscin content in the human macula.
Materials and Methods
Mice were from established colonies at the Medical University of South Carolina, and reared in cyclic light (6:00 A.M. to 18:00 P.M.). The light intensity at cage level during the day part of the cycle was 130–170 lux. Wild type C57BL/6 mice were originally obtained from Harlan Laboratories (Indianapolis, IN); original Nrl−/− breeding pairs, on a C57BL/6 background, were kindly provided by Dr. Anand Swaroop. Animal ages were 1–12 months. For experiments, animals were dark-adapted overnight prior to sacrifice. All protocols were approved by the Institutional Animal Care and Use Committee of the Medical University of South Carolina and followed recommendations from the Panel on Euthanasia of the American Veterinary Medical Association.
Eye fixation, sectioning and immunostaining procedures were as described previously (26). The PETLET polyclonal rabbit anti-RPE65 antibody was used at a concentration of 2 μg mL−1. The short-wavelength S-cone opsin antibody was obtained from Santa Cruz Biotechnology, Inc. (Cat# sc-14363; Santa Cruz, CA), and was used at a concentration of 0.4 μg mL−1. Images were acquired by confocal microscopy (Leica TCS SP5 AOBS Confocal Microscope System; Leica Microsystems, Exton, PA).
In preparation for fluorescence imaging, eyes were cleaned of fat and muscle, then hemisected at the ora serrata, whereas submerged in mammalian physiological buffer (in mm: 130 NaCl, 5 KCl, 0.5 MgCl2, 2 CaCl2, 25 hemisodium-HEPES, 5 glucose, pH = 7.40). After removal of the lens, vitreous and retina, four incisions were made in the eyecup, which was then flattened on a glass slide with the RPE facing up and gently covered with a glass coverslip.
Color photographs were taken with a Nikon D200 digital camera (Nikon, Inc., Melville, NY) through a Zeiss Axioplan 2 microscope (Carl Zeiss, Thornwood, NY) using a 63× oil immersion objective (NA = 1.4). Fluorescence excitation light was 450–490 nm and the emission was collected >510 nm.
For lipofuscin fluorescence measurements, eyecups were imaged on a SP2 Leica laser scanning confocal microscope (Leica Microsystems) with a 10× lens (NA = 0.3) using 488 nm excitation, with the pinhole fully open, and emission collected from 565–725 nm. Fluorescence was measured from a 3.2 × 3.2 mm square around the optic nerve, excluding outer areas of the RPE, which may have been damaged during dissection. The excluded areas have low levels of lipofuscin (<20% of total) and their exclusion does not impact the conclusions of this study. Background fluorescence was corrected for by subtracting the fluorescence of an area adjacent to the eyecup, which did not contain tissue. To allow comparisons across eyes and account for varying detector settings, fluorescence intensity for each eyecup was measured in “bead units” (BU) using InSpeckOrange (540/560) Microscope Image Calibration Kit beads (Molecular Probes/Life Technologies, Carlsbad, CA). Eyecup lipofuscin fluorescence is reported as “BU per Megapixel,” (BU/MP), with one megapixel being equal to 2.44 mm2. Six eyecups from three animals were used for each age.
Lipofuscin granule spectra were measured from 515 to 739 nm (bandwidth of 14 nm) with excitation wavelength of 488 nm using a 63× oil immersion lens (NA = 1.4). To reduce contribution from surrounding tissue fluorescence, the pinhole aperture was reduced to ca 1 Airy. Background fluorescence was corrected for by subtracting the spectrum of an area of the same field, which did not contain granules.
For A2E measurements, after enucleation, hemisection and removal of the lens, vitreous and retina, the RPE-choroid were gently scraped from the sclera and collected in mammalian physiological buffer. The solution was centrifuged in a tabletop centrifuge (16 000 g; Eppendorf Centrifuge 5415C; Eppendorf, AG, Hamburg, Germany) to remove excess buffer. The RPE-choroid tissue pellets were homogenized in phosphate-buffered saline and 1:1 chloroform–methanol, followed by one wash with chloroform and one with methylene chloride. Supernatants from homogenizations and washes were combined, vortexed and centrifuged. The organic solvent layer was carefully collected and dried under argon. Two to six eyecups were used per experiment. Each dried sample was resuspended in methanol with 0.1% trifluoroacetic acid (TFA) and analyzed with a Waters 1525 Binary HPLC (Waters Corporation, Milford, MA) using a reverse-phase gradient from 85% acetonitrile and 15% water with to 100% acetonitrile. Acetonitrile and water used for reverse-phase HPLC contained 0.1% TFA as an ion pairing agent. The A2E peak was ascertained by absorbance spectrum and retention time in comparison to a synthetic A2E standard (8), measured by a Waters 2998 PDA detector. The amount of A2E present was determined from the area under the curve. The peak was collected and A2E identification and quantification were confirmed by LC-MS/MS (27). Experiments were in triplicate. All reagents were of analytical grade; organic solvents were HPLC grade.
Results and Discussion
The photoreceptors of Nrl−/− mice are known to share many properties with cones (24). Figure 1A demonstrates the known characteristic rosette pattern in these animals (23,24). Figure 1B demonstrates the presence of mouse cone visual pigment S-opsin in their outer segments. Staining is found in the rosettes of the outer nuclear layer of Nrl−/− retinas that have been characterized previously (23,24). Interestingly, and confirming the cone-like nature of Nrl−/− photoreceptors, Fig. 1C shows that they also contain the RPE65 protein, which has recently been shown to be present within cones of the C57BL/6 mouse (26) and human retina (28). Figure 1D shows the colocalization of the S-opsin cone pigment and RPE65, suggesting that RPE65 is present within the outer segments of Nrl−/− mouse cells. Therefore, the photoreceptors of the Nrl−/− mice are indeed of cone origin and provide a model for the role of cone photoreceptors in A2E and lipofuscin accumulation. Recent study has shown the accumulation of lipofuscin granules and A2E in the RPE of Nrl−/− mice (29).
Fluorescence photographs of Nrl−/− RPE (Fig. 2B) show distinctive golden-orange granules characteristic of lipofuscin deposits (3). The appearance is similar to those in wild type RPE (Fig. 2A), and indeed the fluorescence emission spectrum of lipofuscin in both strains is a broad peak with emission maximum ca 610 nm (Fig. 2C). The spectrum did not depend on the age of the animals. Lipofuscin granules were smaller and the fluorescence intensity was much lower in Nrl−/− RPE than in RPE of wild type rod-dominant mice, broadly consistent with the previous report (29). The lower granule fluorescence intensity compared to the rest of the tissue probably allowed more relative contribution of background tissue fluorescence in the granule spectra, which would explain the slightly broader and shifted emission spectrum for Nrl−/− mice (Fig. 2C). Another possibility, but for which we have no independent evidence, is that the different fluorescence emission spectra are due to differences in lipofuscin composition between the two mouse strains.
A2E was present in modest amounts in the RPE of Nrl−/− mice (Fig. 3) as shown by analysis of chloroform extracts of Nrl−/− RPE-choroid samples with reverse-phase HPLC and UV–VIS absorption spectroscopy. The presence of A2E was confirmed with liquid chromatography tandem mass spectrometry, as the relevant chromatographic fraction from the Nrl−/− RPE extracts displayed the characteristic fragmentation pattern of A2E, with major peaks at 392, 404, 418, 442 and 486 m/z. Data from a synthetic standard are shown for comparison.
Both lipofuscin fluorescence and A2E levels increased with age in the RPE of Nrl−/− mice (Fig. 4). A2E increased from 0.7 ± 0.3 pmol per eye at 1 month, and at a rate of 0.10 ± 0.04 pmol per eye per month, to reach a level of 2.0 ± 0.8 pmol per eye at 12 months. This is substantially lower than the level of 9.6 ± 0.9 pmol per eye determined in the RPE from 12-month old C57BL/6 mice. Lipofuscin fluorescence was also lower, 26.2 ± 1.1 BU per MP in eyecups of 12-month old Nrl−/−, compared with 171.3 ± 10.6 BU per MP in eyecups of 12-month old C57BL/6 mice. The A2E levels are lower than those recently reported for Nrl−/− (29). These discrepancies may possibly be due to the inclusion of the retina tissue in the previous study, or perhaps to different methods of quantification. The lower levels of lipofuscin and A2E in the Nrl−/− mice compared with wild type animals are not unreasonable, as the total visual pigment present in Nrl−/− retinas is ca 15% of wild type levels (24). Another possibility for the lower levels could be defective phagocytosis (30), which would result in less amounts of outer segment material being digested by the RPE. Such a defect has been attributed to the presence of rosettes in Nrl−/− retinas (Fig. 1), the photoreceptors of which lack access to the RPE. Finally, the lower levels may be due to general retinal degeneration. Interestingly, Conley et al. (29) have shown that the absence of Abca4 increases the levels of both A2E and lipofuscin in the Nrl−/− cone-like dominant retina and this increase is to a greater extent than in the rod-dominant retinas lacking Abca4.
To summarize, our results show that cone visual pigments and photoreceptors in the Nrl−/− mouse can contribute to lipofuscin and A2E formation and the levels increase with age. Extrapolation of this result to the human macula would suggest that cone photoreceptors may contribute significantly to the formation of lipofuscin in that region. A careful analysis of the composition of lipofuscin from cone-dominant retinas versus lipofuscin from rod-dominant retinas may be informative for understanding the high concentration of lipofuscin in the center of the human RPE.
Acknowledgements— This work was supported by NIH grants EY020661 (ZA/RKC), EY004939 (RKC), and EY014850 (YK); Foundation Fighting Blindness, Inc. (Owings Mills, MD; RKC); and unrestricted awards to the Departments of Ophthalmology at MUSC from Research to Prevent Blindness (RPB; New York); RKC is an RPB Senior Scientific Investigator. This work was conducted in a facility constructed with support from the National Institutes of Health, Grant Number C06 RR015455 from the Extramural Research Facilities Program of the National Center for Research Resources.
- 221998) The First Steps in Seeing, pp. 447–452. Sinauer Associates, Inc, Sunderland, MA.(