You are here

Article Text

Article menu
Laboratory science
High-resolution imaging of autofluorescent particles within drusen using structured illumination microscopy
  1. Sabrina Rossberger1,2,
  2. Thomas Ach1,
  3. Gerrit Best1,2,
  4. Christoph Cremer2,3,
  5. Rainer Heintzmann4,5,6,
  6. Stefan Dithmar1
  1. 1Department of Ophthalmology, University of Heidelberg, Heidelberg, Germany
  2. 2Kirchhoff Institute for Physics, University of Heidelberg, Heidelberg, Germany
  3. 3Institute of Molecular Biology (IMB), Mainz, Germany
  4. 4Institute for Physical Chemistry, Abbe Center of Photonics, Friedrich Schiller University Jena, Jena, Germany
  5. 5Institute of Photonic Technology, Jena, Germany
  6. 6Randall Division, King's College London, London, UK
  1. Correspondence to Professor Stefan Dithmar, Department of Ophthalmology, University of Heidelberg, Im Neuenheimer Feld 400, Heidelberg 69120, Germany; stefan.dithmar{at}med.uni-heidelberg.de

Abstract

Purpose Autofluorescent (AF) material within drusen has rarely been described and there is little knowledge about origin and formation of these particles. Drusen formation is still a relatively unknown process and analysis of AF inclusions might be important for the understanding of fundamental processes. Here we present a detailed analysis of drusen containing AF material using structured illumination microscopy (SIM), which provides a lateral resolution twice as high as conventional fluorescence microscopy.

Methods Eight histological retinal pigment epithelium (RPE) sections obtained from eight human donor eyes (76±4 years) were examined by SIM using laser light of different wavelengths (488 nm, 568 nm). Drusen were studied with regards to their size and shape. AF material within drusen was analysed in terms of size, shape, AF behaviour, and distribution across drusen.

Results A total of 441 drusen were found, of which 101 contained AF material (22.9%). 90.1% of these drusen were smaller than 63 µm (mean: 35.65 µm±2.38 µm) regardless of whether classified as hard or soft drusen. AF particles (n=190) within drusen show similar spectra compared with lipofuscin granules in RPE cells. Up to 11 particles were found within a single druse. Nearly all particles were located in the outer 2/3 of the drusen (85.94%).

Conclusions SIM allows studying AF particles within drusen on a higher resolution level compared with conventional fluorescence, multiphoton or even confocal microscopy and therefore provides detailed insights in drusen. Shape and autofluorescence analysis of the material embedded in drusen suggest that these particles originate from the overlaying RPE cells.

  • Imaging
  • Retina

https://doi.org/10.1136/bjophthalmol-2012-302350

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Introduction

    Drusen play a key role in age related macular degeneration (ARMD) and appear as focal deposits between the inner collagenous layer of Bruch's membrane (BM) and the basal lamina of the retinal pigment epithelium (RPE).1–4 The pathogenesis of drusen formation is still unknown. They consist of lipids and proteins and occasionally intracellular material like organelles and fragments originating from RPE cells.5–7 There are some reports (histological and in vivo) referring to autofluorescence (AF) of extracted drusen, AF of RPE cells overlying drusen or AF within drusen.8–10 In a few cases, particles/granules within drusen were observed, which consist of strongly AF agents in the visible spectrum and show similarities to lipofuscin granules (LF) of RPE cells.11–13

    Although different imaging methods like conventional light microscopy,14 laser scanning microscopy15 and electron microscopy6 ,7 have been applied to study drusen in detail, there is no data available on frequency, size and behaviour of AF structures within drusen.

    Electron microscopy is useful for examining the composition of drusen, however it cannot assess AF characteristics.6 ,16 Fluorescence microscopy is able to assess AF characteristics, but it is restricted by the resolution limit (wide-field microscopy: ∼230 nm, confocal microscopy: ∼180 nm, two-photon microscopy: ∼200 nm).17–19 We recently introduced structured illumination microscopy (SIM) as a new tool for high-resolution imaging of AF granules within RPE cells.20–22 This method benefits from simple sample preparation and provides a resolution that is twice as good compared with standard wide-field fluorescence microscopy. The resolving power is also superior to confocal23 or multiphoton microscopy.24 Thus SIM allows a much more differentiated analysis of the AF tissue.

    In this study, SIM was used for the detection and characterisation of fluorescent structures within drusen in histological sections.

    Methods

    Sample preparation

    Histological sections from ora serrata to ora serrata traversing the macular region were obtained from eight human donor eyes (76±4 years) from the Department of Ophthalmology, University of Heidelberg. In all donor eyes, no macroscopic retinal abnormalities were detectable except for drusen. There were no signs of advanced dry or exsudative ARMD.

    Immediately after removing the anterior segment of the eyes for corneal transplantation, the remaining tissue was fixed with 4% paraformaldehyde in phosphate-buffered saline (pH 7.4) and embedded in paraffin. Sections (4 µm) were prepared and mounted on microscope slides, deparaffinised with xylene (3×5 min), rehydrated through graded ethanol and stored in phosphate-buffered saline until further processing. For microscopy, tissue sections were embedded with SlowFade antifade (Invitrogen, Carlsbad, California, USA), coverslipped and edges were sealed with nail polish. One section per donor was further examined.

    Approval was obtained from the local ethics committee. All procedures adhered to the Tenets of the Declaration of Helsinki.

    Structured illumination microscopy

    SIM is a special fluorescence microscopy method, which allows lateral resolution twice as high as in commonly used wide-field microscopes, in which resolution is limited by the wavelength of light.25 ,26

    SIM is based on the moiré-effect: two overlaying gratings result in a coarser (moiré) structure, which can be resolved more easily. If the coarse grating is imaged and measured and one of the finer gratings is known, the other fine grating (ie, the sample structure) can be calculated (figure 1). Briefly, the samples in this study were illuminated with a fine sinusoidal pattern (350 nm period), which was shifted by 1/3×350 nm perpendicular to the direction of the grating stripes. Thus three images with different positions of the illumination grating on the sample were obtained (grating phase positions). Note that the sum of the three images corresponds to an image obtained for homogeneous illumination. In order to achieve an isotropic resolution improvement in the lateral plane the three images of changing phases were recorded for each of three orientations of the fine grating (sinusoidal pattern), which was automatically rotated (0°, 60° and 120°). Thus a total of nine images of standard resolution were acquired (three phase images, three in each direction), which were further processed using previously described custom reconstruction software.19 ,20 ,26

    Figure 1
    Figure 1

    Moiré-effect: Overlaying two fine gratings result in a coarse, but resolvable grating. If the coarse grating as well as one of the finer gratings is known, the other fine grating will be calculable.

    A custom setup was used for SIM imaging: two monochromatic excitation wavelengths were used for all experiments—a cyan 488 nm and a yellow 568 nm laser line (Coherent Sapphire 488/568 HP, 200 mW, Coherent, Dieburg, Germany). The AF signal emitted by the sample passed the high numerical objective (Leica HCX PL APO 100x/1.4 oil CS) and was separated from the excitation light by a triple-band bandpass emission filter (Z488/568/647 M V2, AHF, Tübingen, Germany). Afterwards, an image was formed by the tube lenses and detected with a charge-coupled device-camera (Sensicam QE, PCO, Kelheim, Germany). A detailed description of the microscope setup has recently been published.20

    The acquisition software is based on the python programming language.

    Imaging

    To prepare the imaging process with SIM, one histological section of each donor was analysed with regards to drusen, using a commercial epifluorescent microscope (Leica DMRB, Wetzlar, Germany; excitation: 404 nm, camera: Quanticam SVGA-color, Phase, Lübeck, Germany, objective: HCX PL APO 63x/1.4 oil CS). Each druse was imaged independently of whether containing AF inclusions or not.

    The same sections were then studied for drusen with the SIM setup, using low laser intensities compared with standard fluorescence microscopy (SIM excitation intensity: ∼0.17 mW/cm2±6%) in order to avoid prebleaching.

    Each druse was scanned manually for AF particles from the surface to the bottom (in depth, parallel to BM; z-direction). The focal layer with the largest diameter of the fluorescent particle was chosen. Then laser power was switched to higher intensity to perform SIM-image acquisition (excitation intensity: ∼87 mW/cm2±6%; 50 ms/image). At first, the laser with the longer wavelength (568 nm) was used for image acquisition. Immediately afterwards the same druse was imaged using the shorter wavelengths (488 nm). After the image reconstruction process one image for each excitation wavelength was obtained.

    Analysis

    Drusen were classified into hard and soft type based on their shape but not content as sections have been treated with organic solvents during the deparaffinisation process. Drusen appearing dome-shaped with well-defined borders were classified as ‘hard’ type, whereas drusen with poorly defined borders with endings often merging with adjacent basal linear deposits were assigned to the ‘soft’ type.13 Basal diameters were measured using the image processing software ImageJ, (Open Source, National Institute of Health, USA). The number of drusen containing fluorescent material detected by epifluorescence microscopy and SIM was compared.

    SIM images were used for further analysis: fluorescent particles within drusen were measured (greatest diameter) and classified into smaller (<500 nm in diameter) and larger ones (>500 nm in diameter). Their distance to BM was determined.

    To clarify if AF particles located in drusen are similar to intracellular RPE granules overlying the druse, both entities were analysed with regards to spectral similarities. For this purpose, two intensity profiles of selected AF particles and RPE-LF granules were generated (one for each excitation wavelength). Calculating the characteristic ratio of the maxima given by the intensity profiles allows a comparison of RPE-LF and AF particles within drusen.

    Results

    A total of 441 drusen in eight ora serrata to ora serrata histological sections was identified using epifluorescence microscopy. Hard and soft drusen were on average equally distributed among the eight sections (table 1). Fluorescent material within drusen was found in 17.91% of all drusen (n=79) using epifluorescence microscopy (not shown in table) and in 22.68% (n=100) using SIM.

    View this table:
    Table 1

    Characteristics of drusen of eight ora to ora histological sections traversing the macular region using epifluorescence microscopy and structured illumination microscopy (SIM)

    Fifty-five per cent of drusen with embedded AF material and studied with SIM were assigned to the ‘hard’ type (table 1, figure 2), whereas the remaining 45% were classified as ‘soft’ drusen (table 1, figure 3). In concordance with the histological classification of drusen by Rudolf et al13 drusen size and shape are not necessarily related. Hard drusen (basal diameter: 29.40 µm (mean) ±19.88 µm (SD)) were on average smaller than soft drusen, however the majority of soft drusen measured less than 63 µm (43.59 µm± 26.25 µm), 90% of all drusen with embedded fluorescent material were smaller than 63 µm (mean: 35.79 µm± 23.97 µm; table 2).27

    View this table:
    Table 2

    Drusen imaged with structured illumination microscopy  and found to contain fluorescent particles were distinguished into hard and soft drusen.13 Additionally, a size classification according to the Wisconsin grading system was conducted27

    Figure 2
    Figure 2

    Typical autofluorescent particle (arrow) found within a hard druse smaller than 63 μm (size of druse: 31.08 μm (arrowheads); BM, Bruch's membrane; LF, lipofuscin; MLF, melanolipofuscin; RPE, retinal pigment epithelium; scale bar: 2 μm; red channel: 488 nm excitation; green channel: 568 nm excitation).

    Figure 3
    Figure 3

    Typical autofluorescent particle (arrow) found within a soft druse smaller than 63 μm (size of druse: 47.35 μm (arrowheads); BM, Bruch's membrane; LF, lipofuscin; MLF, melanolipofuscin; RPE, retinal pigment epithelium; scale bar: 2 μm).

    Autofluorescence intensity measurements allowed comparing the signal ratio of the fluorescent material found in drusen with the signals of LF granules deposited in RPE cells overlaying the druse, which is exemplarily demonstrated in figure 4. Here, the intensity profiles for each wavelength showed marked similarities with regards to shape and maximum intensity (figure 4) when comparing a RPE-LF granule with an AF particle embedded in the druse. Autofluorescence intensity of the RPE-LF granule was higher at 488 nm (max. intensity: 213 a.u.) than at 568 nm (max. intensity: 45 a.u.). The ratio of the maximum intensity at 568 nm excitation compared with 488 nm excitation was 0.21. A similar intensity profile and ratio was observed for the AF particle embedded in the druse (max. intensity at 488 nm: 203 a.u.; at 568 nm: 42 a.u.; intensity ratio 568/488=0.21). These similarities have been observed for other AF particles within drusen as well.

    Figure 4
    Figure 4

    Autofluorescence profiles of a particle within a druse and RPE-LF. The comparison of profiles shows marked similarities in terms of fluorescence properties and shape. (A) Typical autofluorescent (AF) particles found within a soft druse (size of druse: 48.18 µm; scale bar: 2 µm; red channel: 488 nm excitation; green channel: 568 nm excitation). (B) Intensity profile of a LF granule within a RPE cell overlaying the druse (white circle; z: intensity (a.u.); B1: 488 nm excitation, max.: 213 a.u.; B2: 568 nm excitation, max.: 45 a.u.). (C) Intensity profile of an AF particle within a druse (red circle; z: intensity (a.u.); C1: 488 nm excitation, max.: 203 a.u.; C2: 568 nm excitation, max.: 42 a.u.).

    A total of 189 AF particles were found in 100 drusen. Single drusen contained 1–11 particles. Some particles appeared like LF granules in RPE cells. Others were smaller than RPE-LF granules and a few lacked the characteristic round shape, though the intensity ratio given by the two excitation wavelengths was similar to LF granules of the RPE (figure 5). AF particles within drusen had a median size of 702 nm±297 nm. Fifty-seven (30.16%) particles were smaller than 500 nm (figure 6). Of all AF particles 83.07% were located in the outer two-third between BM and RPE. On average, the distance between particle and BM was 6.31 µm± 7.12 µm. In comparison, the heights of the drusen averaged out at 11.73 µm± 6.87 µm.

    Figure 5
    Figure 5

    Autofluorescence profiles of particles smaller than 500 nm within a druse (A, B, C; z: intensity (a.u.)). The small particles show similar fluorescence properties as RPE-LF granules (D; z: intensity (a.u.)) though size and shape lack the typical round-shaped appearance (left and right row). As particles are smaller, maximum intensities emitted (max.: ∼100 a.u., middle row: A–C) were generally lower compared with RPE-LF (max.: 180 a.u., middle row: D) presumably due to less LF accumulation (size of druse: 66.88 µm; scale bar:  2 µm; magnification: 300 nm; excitation: 488 nm).

    Figure 6
    Figure 6

    Size distribution of autofluorescent (AF) particles within drusen: 30.16% of all particles were smaller than 500 nm.

    Discussion

    The pathogenesis of drusen, a major risk factor for ARMD, is still not completely understood. Drusen have been classified as hard and soft drusen and consist of several different components mainly verified in immunohistological studies.11 Observing soft drusen smaller than 63 µm is not in agreement with the general assumption.

    Our result matches with the results presented by Rudolf et al,13 who also reported similar median sizes for hard and soft drusen. Earlier, Hageman and Mullins had already suggested that a ‘strict relationship between size and shape’ does not exist.7 However, the range covered by soft drusen is still much larger (maximum size found: 130 µm) as expected. Nevertheless, it has to be mentioned, that the data presented in this study is only referring to one section of each donor—thus we cannot provide absolute numbers of drusen in the whole eye and sections might not represent the whole donor eye.

    Intracellular material like organelles and fragments presumably originating from RPE cells has been described within drusen and basal laminar deposits.5 ,6 ,11 ,28 While some studies report such structures as coincidental findings,7 ,11 ,12 Rudolf et al13 found pigment granules within drusen in frequencies of 6.4% to 8.9%. However, little is known about frequency, shape and fluorescent properties of structures embedded in drusen.

    SIM is a new microscopic method, which allows high-resolution fluorescent imaging. SIM provides benefits of a resolution twice as good as standard fluorescence microscopy25 ,26 and also provides a better resolution compared with confocal and multiphoton microscopy.23 ,24 The main advantage of the resolution improvement achieved with SIM emerges when analysing the AF particles in terms of size, shape and intensity profiles. Here, more detailed analyses are possible compared with standard wide-field microscopy. Using standard wide-field microscopy, 79 drusen containing fluorescent material were detectable (17.91%), whereas 100 drusen with fluorescent particles were found using SIM (22.68%). AF particles were almost equally present in hard and soft drusen, thus drusen shape does not seem to be important for the presence of AF structures. Drusen size however seems to matter: 90% of all drusen with embedded AF particles were smaller than 63 µm. Autofluorescence intensity measurements and comparison of intensity ratios for different wavelengths allowed drawing the conclusion that fluorescent material in drusen resembles LF granules of the RPE. Similar ratios are a significant indication that AF particles and RPE-LF granules are composed of the same material though no whole emission spectrum was available. As intensity profiles of AF particles were always compared with RPE-LF granules within cells overlaying the druse (on the same section), possible changes by the fixation process on the AF characteristics will have affected same materials in the same manner. Thus a misinterpretation of the spectral signals can be excluded as AF particles were always compared with RPE-LF granules in the proximate vicinity.

    AF particle intensity profiles showed a stronger fluorescence for 488 nm compared with 568 nm excitation as expected for LF granules. Furthermore the fluorescent particles in drusen generally appeared round-shaped as RPE-LF granules and were well circumscribed with no signs of dissociation.

    A detailed analysis of particle sizes yielded a mean of 702 nm±297 nm. This value is smaller than the typical size of LF granules within RPE cells, which have previously been evaluated by Boulton et al,29 who demonstrated a LF size of about 1–3 µm using electron transmission microscopy. Biesemeier et al30 have recently reported a size of 1.5 µm×0.5 µm for granules within RPE cells using electron microscopy also but without distinguishing between LF, melanin and melanolipofuscin. We observed that almost one-third (30.16%) of AF particles in drusen were even smaller than 500 nm. This 500 nm threshold is a arbitrarily selected value in this study and far below the previously reported granule sizes in literature. This emphasises the advantage of SIM. SIM is especially useful for the detection of such small particles and low autofluorescence signals. It has to be mentioned that some granules have been observed marginal to the section's thickness. A shortcoming of measuring structures on histological sections generally is that sections are only of a certain thickness, which may result in cropped structures at the edges.

    It has been postulated that inflammatory, cell-mediated and immune processes might initiate drusen formation.11 ,12 Injuries of the RPE caused by various events (gene mutation, light damage, oxidative stress, LF accumulation)11 are assumed to result in a release and accumulation of cell debris between BM and RPE.31 Recently, Johnson et al32 have shown that even cultured RPE cells secrete several membrane-bounded and non-membrane-bounded deposits which are also found in drusen. Besides this, material originating from the RPE, choroid and plasma proteins related to inflammatory processes have been reported within drusen,11 ,33 which supports the idea that drusen biosynthesis is a response to local inflammatory events.34 Moreover electron micrograph images of ‘vacuoles’ containing LF located within the RPE monolayer were interpreted by Anderson et al12 as degenerated RPE cells and the first stage of forming a druse.

    As drusen have been identified to contain lysosomal material, observing very small granules within drusen (<500 nm) could display early stages of catabolic lysosomal processes of RPE cell material within drusen resulting in LF-like deposits.35 Alternatively, lysosomal-LF granules secreted from RPE cells might undergo further catabolic degradation processes resulting in smaller LF particles than originally exocytosed.

    We mainly detected these LF granules within small drusen (<63 µm), which might indicate this aforementioned process of LF appearance only in early stages of drusen formation. In large drusen, LF granules are presumably completely catabolised. However, the AF characteristic must get lost during the suggested catabolic processes, as no fragments in the near vicinity of a granule were observable. It has to be mentioned that though AF particles were mainly detected within small drusen, no correlation between drusen volume and AF particle volume was observed (cf. see online supplementary material).

    Furthermore, nearly all AF LF granules (83.07%) have been found in the outer two-third of the druse. This also might hint at secretion of AF deposits in early stages of drusen formation. Or it indicates the time-consuming process of LF-like granules formulation within a druse while further excretion of RPE cell material is ongoing.

    The data presented in this study is in concordance with the hypotheses of drusen originating partly from RPE cell material. SIM has proved to be a useful tool for analysing fluorescent particles at a resolution level, which is much better than standard fluorescence microscopy and therefore provides better insights in drusen formation.

    References

    1. ↵
      1. Pauleikhoff D,
      2. Barondes MJ,
      3. Minassian D,
      4. et al
      . Drusen as risk factors in age-related macular disease. Am J Ophthalmol 1990;109:38–43.
      OpenUrlPubMedWeb of Science
    2. ↵
      1. Abdelsalam A,
      2. Del Priore L,
      3. Zarbin MA
      . Drusen in age-related macular degeneration: pathogenesis, natural course, and laser photocoagulation-induced regression. Surv Ophthalmol 1999;44:1–29.
      OpenUrlCrossRefPubMedWeb of Science
    3. ↵
      1. Curcio CA,
      2. Millican CL
      . Basal linear deposit and large drusen are specific for early age-related maculopathy. Arch Ophthalmol 1999;117:329–39.
      OpenUrlCrossRefPubMedWeb of Science
    4. ↵
      1. Bird AC,
      2. Bressler NM,
      3. Bressler SB,
      4. et al
      . An international classification and grading system for age-related maculopathy and age-related macular degeneration. The International ARM Epidemiological Study Group. Surv Ophthalmol 1995;39:367–74.
      OpenUrlCrossRefPubMedWeb of Science
    5. ↵
      1. Curcio CA,
      2. Medeiros NE,
      3. Millican CL
      . The alabama age-related macular degeneration grading system for donor eyes. Invest Ophthalmol Vis Sci 1998;39:1085–96.
      OpenUrlAbstract/FREE Full Text
    6. ↵
      1. Wang L,
      2. Clark ME,
      3. Crossman DK,
      4. et al
      . Abundant lipid and protein components of drusen. PLoS One 2010;5:e10329.
      OpenUrlCrossRefPubMed
    7. ↵
      1. Hageman GS,
      2. Mullins RF
      . Molecular composition of drusen as related to substructural phenotype. Mol Vis 1999;5:28.
      OpenUrlPubMed
    8. ↵
      1. Newsome DA,
      2. Hewitt AT,
      3. Huh W,
      4. et al
      . Detection of specific extracellular matrix molecules in drusen, Bruch's membrane, and ciliary body. Am J Ophthalmol 1987;104:373–81.
      OpenUrlPubMedWeb of Science
    9. ↵
      1. Marmorstein AD,
      2. Marmorstein LY,
      3. Sakaguchi H,
      4. et al
      . Spectral profiling of autofluorescence associated with lipofuscin, Bruch's Membrane, and sub-RPE deposits in normal and AMD eyes. Invest Ophthalmol Vis Sci 2002;43:2435–41.
      OpenUrlAbstract/FREE Full Text
    10. ↵
      1. Delori FC,
      2. Fleckner MR,
      3. Goger DG,
      4. et al
      . Autofluorescence distribution associated with drusen in age-related macular degeneration. Invest Ophthalmol Vis Sci 2000;41:496–504.
      OpenUrlAbstract/FREE Full Text
    11. ↵
      1. Hageman GS,
      2. Luthert PJ,
      3. Chong NH Victor,
      4. et al
      . An integrated hypothesis that considers drusen as biomarkers of immune-mediated processes at the RPE-Bruch's membrane interface in aging and age-related macular degeneration. Prog Retin Eye Res 2001;20:705–32.
      OpenUrlCrossRefPubMedWeb of Science
    12. ↵
      1. Anderson DH,
      2. Mullins RF,
      3. Hageman GS,
      4. et al
      . A role for local inflammation in the formation of drusen in the aging eye. Am J Ophthalmol 2002;134:411–31.
      OpenUrlCrossRefPubMedWeb of Science
    13. ↵
      1. Rudolf M,
      2. Clark ME,
      3. Chimento MF,
      4. et al
      . Prevalence and morphology of druse types in the macula and periphery of eyes with age-related maculopathy. Invest Ophthalmol Vis Sci 2008;49:1200–9.
      OpenUrlAbstract/FREE Full Text
    14. ↵
      1. Rudolf M,
      2. Malek G,
      3. Messinger JD,
      4. et al
      . Sub-retinal drusenoid deposits in human retina: organization and composition. Exp Eye Res 2008;87:402–8.
      OpenUrlCrossRefPubMedWeb of Science
    15. ↵
      1. Spaide RF,
      2. Curcio CA
      . Drusen characterization with multimodal imaging. Retina 2010;30:1441–54.
      OpenUrlCrossRefPubMedWeb of Science
    16. ↵
      1. Gouras P,
      2. Ivert L,
      3. Mattison JA,
      4. et al
      . Drusenoid maculopathy in rhesus monkeys: autofluorescence, lipofuscin and drusen pathogenesis. Graefes Arch Clin Exp Ophthalmol 2008;246:1403–11.
      OpenUrlCrossRefPubMed
    17. ↵
      1. Abbe E
      . Beitraege zur Theorie des Mikroskops und der mikroskopischen Wahrnehmung. Arch Mikrosk Anat 1873;9:413–20.
      OpenUrlCrossRef
    18. ↵
      1. Rayleigh L
      . On the theory of optical images with special reference to the microscope. Philos Mag 1896;42:167–95.
      OpenUrl
    19. ↵
      1. Heintzmann R,
      2. Ficz G
      . Breaking the resolution limit in light microscopy. Brief Funct Genomic Proteomic 2006;5:289–301.
      OpenUrlAbstract/FREE Full Text
    20. ↵
      1. Best G,
      2. Amberger R,
      3. Baddeley D,
      4. et al
      . Structured illumination microscopy of autofluorescent aggregations in human tissue. Micron 2011;42:330–5.
      OpenUrlCrossRef
    21. ↵
      1. Ach T,
      2. Best G,
      3. Ruppenstein M,
      4. et al
      . High-resolution fluorescence microscopy of retinal pigment epithelium using structured illumination. Ophthalmologe 2010;107:1037–42.
      OpenUrlCrossRefPubMed
    22. ↵
      1. Ach T,
      2. Best G,
      3. Rossberger S,
      4. et al
      . Autofluorescence imaging of human RPE cell granules using structured illumination microscopy. Br J Ophthalmol 2012;96:1141–4.
      OpenUrlAbstract/FREE Full Text
    23. ↵
      1. Schermelleh L,
      2. Heintzmann R,
      3. Leonhardt H
      . A guide to super-resolution fluorescence microscopy. J Cell Biol 2010;190:165–75.
      OpenUrlAbstract/FREE Full Text
    24. ↵
      1. Denk W,
      2. Strickler JH,
      3. Webb WW
      . Two-photon laser scanning fluorescence microscopy. Science 1990;248:73–6.
      OpenUrlAbstract/FREE Full Text
    25. ↵
      1. Gustafsson MG
      . Surpassing the lateral resolution limit by a factor of two using structured illumination microscopy. J Microsc 2000;198(Pt 2):82–7.
      OpenUrlCrossRefPubMedWeb of Science
    26. ↵
      1. Heintzmann R,
      2. Cremer C
      . Laterally modulated excitation microscopy: improvement of resolution by using a diffraction grating. Proc SPIE 1999;3568:185–96.
      OpenUrlCrossRef
    27. ↵
      1. Klein R,
      2. Davis MD,
      3. Magli YL,
      4. et al
      . The Wisconsin age-related maculopathy grading system. Ophthalmology 1991;98:1128–34.
      OpenUrlCrossRefPubMedWeb of Science
    28. ↵
      1. Vogt SD,
      2. Curcio CA,
      3. Wang L,
      4. et al
      . Retinal pigment epithelial expression of complement regulator CD46 is altered early in the course of geographic atrophy. Exp Eye Res 2011;93:413–23.
      OpenUrlCrossRefPubMedWeb of Science
    29. ↵
      1. Boulton M,
      2. McKechnie NM,
      3. Breda J,
      4. et al
      . The formation of autofluorescent granules in cultured human RPE. Invest Ophthalmol Vis Sci 1989;30:82–9.
      OpenUrlAbstract/FREE Full Text
    30. ↵
      1. Biesemeier A,
      2. Schraermeyer U,
      3. Eibl O
      . Chemical composition of melanosomes, lipofuscin and melanolipofuscin granules of human RPE tissues. Exp Eye Res 2011;93:29–39.
      OpenUrlCrossRefPubMed
    31. ↵
      1. Burns RP,
      2. Feeney-Burns L
      . Clinico-morphologic correlations of drusen of Bruch's membrane. Trans Am Ophthalmol Soc 1980;78:206–25.
      OpenUrlPubMed
    32. ↵
      1. Johnson LV,
      2. Forest DL,
      3. Banna CD,
      4. et al
      . Cell culture model that mimics drusen formation and triggers complement activation associated with age-related macular degeneration. Proc Natl Acad Sci USA 2011;108:18277–82.
      OpenUrlAbstract/FREE Full Text
    33. ↵
      1. Johnson PT,
      2. Lewis GP,
      3. Talaga KC,
      4. et al
      . Drusen-associated degeneration in the retina. Invest Ophthalmol Vis Sci 2003;44:4481–8.
      OpenUrlAbstract/FREE Full Text
    34. ↵
      1. Johnson LV,
      2. Leitner WP,
      3. Staples MK,
      4. et al
      . Complement activation and inflammatory processes in Drusen formation and age related macular degeneration. Exp Eye Res 2001;73:887–96.
      OpenUrlCrossRefPubMedWeb of Science
    35. ↵
      1. Chen PM,
      2. Gombart ZJ,
      3. Chen JW
      . Chloroquine treatment of ARPE-19 cells leads to lysosome dilation and intracellular lipid accumulation: possible implications of lysosomal dysfunction in macular degeneration. Cell Biosci 2011;1:10.
      OpenUrlCrossRefPubMed

    Supplementary materials

    • Supplementary Data

      This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

      Files in this Data Supplement:

    Footnotes

    • Contributors All authors have made substantive contributions to this study.

      SR: analysis and interpretation of data, drafting the article, final approval of the version to be published. TA: conception and design, critically revising the article, analysis and interpretation of data, final approval of the version to be published. GB: analysis and interpretation of data, drafting the article, final approval of the version to be published. RH: substantial contributions to conception, critically revising the article, final approval of the version to be published. CC: substantial contributions to conception, critically revising the article, final approval of the version to be published. SD: analysis and interpretation of data, critically revising the article, final approval of the version to be published.

    • Funding This study was supported by Ernst and Berta Grimmke Stiftung (No. 01/10). The funding source had no involvement in the study design, the collection, analysis and interpretation of data, in the writing of the report or in the decision to submit the paper for publication.

    • Competing interests None.

    • Ethics approval Ethics Committee University of Heidelberg.

    • Provenance and peer review Not commissioned; externally peer reviewed.