SEARCH

SEARCH BY CITATION

Keywords:

  • aging;
  • lipofuscin;
  • microglia;
  • retina

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References

Fundus autofluorescence (AF) imaging by confocal scanning laser ophthalmoscopy has been widely used by ophthalmologists in the diagnosis/monitoring of various retinal disorders. It is believed that fundus AF is derived from lipofuscin in retinal pigment epithelial (RPE) cells; however, direct clinicopathological correlation has not been possible in humans. We examined fundus AF by confocal scanning laser ophthalmoscopy and confocal microscopy in normal C57BL/6 mice of different ages. Increasingly strong AF signals were observed with age in the neuroretina and subretinal/RPE layer by confocal scanning laser ophthalmoscopy. Unlike fundus AF detected in normal human subjects, mouse fundus AF appeared as discrete foci distributed throughout the retina. Most of the AF signals in the neuroretina were distributed around retinal vessels. Confocal microscopy of retinal and choroid/RPE flat mounts demonstrated that most of the AF signals were derived from Iba-1+ perivascular and subretinal microglia. An age-dependent accumulation of Iba-1+ microglia at the subretinal space was observed. Lipofuscin granules were detected in large numbers in subretinal microglia by electron microscopy. The number of AF+ microglia and the amount of AF granules/cell increased with age. AF granules/lipofuscin were also observed in RPE cells in mice older than 12 months, but the number of AF+ RPE cells was very low (1.48 mm−2 and 5.02 mm−2 for 12 and 24 months, respectively) compared to the number of AF+ microglial cells (20.63 mm−2 and 76.36 mm−2 for 6 and 24 months, respectively). The fluorescence emission fingerprints of AF granules in subretinal microglia were the same as those in RPE cells. Our observation suggests that perivascular and subretinal microglia are the main cells producing lipofuscin in normal aged mouse retina and are responsible for in vivo fundus AF. Microglia may play an important role in retinal aging and age-related retinal diseases.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References

Lipofuscin (from Greek lipo for ‘fat’ and Latin fuscus for ‘dark’) is a chemically and morphologically polymorphous waste material, originating from a variety of intracellular structures, which accumulate at the primary site of waste disposal, the lysosome, of most aging eukaryotic cells (Brunk & Terman, 2002). Lipofuscin is brownish yellow in appearance by conventional microscopy, and contains several fluorophores detectable across a broad spectrum by fluorescence microscopy (Terman & Brunk, 1998; Gray & Woulfe, 2005). Lipofuscin has been considered as a hallmark of aging because of the fact that the amount of lipofuscin increases with age (Strehler et al., 1959). Excessive accumulation of lipofuscin could compromise essential cell function (Brunk & Terman, 2002) and, therefore, contribute to many age-related diseases including Alzheimer's disease (Dowson, 1982) and age-related macular degeneration (AMD) (Dorey et al., 1989; Wolf, 2003; Sparrow & Boulton, 2005).

In the eye, retinal tissue can be directly visualized non-invasively by confocal scanning laser ophthalmoscopy (cSLO) (Woon et al., 1992). An autofluorescence (AF) signal, derived predominantly from lipofuscin, can be obtained using the cSLO with an excitation wavelength of 488 nm and a band-pass filter at 521 nm (von Ruckmann et al., 1995; Bellmann et al., 2003; Bindewald et al., 2005). Autofluorescence imaging information collected in this way can be quantified by the strength of the signal (Lois et al., 2000). Quantitative and qualitative changes in fundus AF are associated with certain fundus diseases (Lois et al., 2000, 2002, 2004; Scholl et al., 2004; Bindewald et al., 2005), and with experience these changes are proving to be of diagnostic and prognostic value.

Because of the fact that retinal pigment epithelial (RPE) cells form and accumulate lipofuscin during aging (see review by Kennedy et al., 1995), it is believed that fundus AF is derived from lipofuscin in RPE cells. Fundus AF has therefore been used clinically to evaluate directly the ‘in vivo’ state of RPE cells and indirectly those areas of the fundus at greatest risk of pathology because of RPE dysfunction. However, the correlation between the fundus AF and RPE lipofuscin has not been histopathologically validated. In this study, we examined fundus AF in normal young and aged mice in vivo using a custom-built cSLO. Eyes were then examined ex vivo for lipofuscin by confocal microscopy of retinal and RPE–choroidal flat mounts, and a clinicopathologic correlation was then sought. An age-dependent increase in fundus AF was observed in experimental mice. Interestingly, AF signals in mouse fundus appeared as discrete high autofluorescent foci localized either around retinal vessels or at the subretinal/RPE regions. Surprisingly, pathological studies revealed that the major proportion of the fundus AF signal appeared to be derived from lipofuscin granules in perivascular and subretinal microglia, rather than from RPE cells.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References

Fundus AF in normal aged mice

Fundus AF was imaged in vivo using a customized high-sensitivity cSLO (Manivannan et al., 1993), with which neuroretinal tissue and RPE/choroid can be imaged separately by adjusting the confocal aperture (Xu et al., 2002). Patches of low AF signals were observed in the fundi of young (3-month-old) mice (Fig. 1A). Fundus AF signals increased with age (Fig. 1). In 12-month-old mice, a few high-AF areas were detected in the deep retinal/RPE layer (Fig. 1B). In addition, patches of low AF were also observed at this layer (Fig. 1B). In mice older than 18 months old, many small high AF foci were detected around retinal vessels (Fig. 1C). In the deep retinal/pre-RPE layer, many more high-AF changes were seen scattered throughout the entire fundus (Fig. 1D). It should be noted, however, that the appearance of mouse fundus AF differs from that usually seen in aged normal human subjects.

image

Figure 1. In vivo fundus autofluorescence (AF) in different ages of mice. Mouse fundus AF was imaged by a homebuilt confocal scanning laser ophthalmoscopy (cSLO) with 488 nm excitation and a 520 filter. (A) Fundus AF from a 3-month-old mouse; cSLO was focused on the retinal pigment epithelial (RPE). Mild background AF is present interrupted by areas of low AF. (B) Fundus AF from a 12-month-old mouse; cSLO was focused on the RPE layer. In addition to the mentioned findings, few bright foci of very high AF signal are also observed. (C and D) Fundus AF from 18-month-old mice. (C) SLO was focused on retinal layer. Multiple foci with very high AF signal are seen surrounding large retinal vessels. (D) SLO was focused on the RPE layer. Multiple foci of very high AF signal are present. OD, optic disk.

Download figure to PowerPoint

Lipofuscin in retinal perivascular and subretinal microglia

To identify the cell source of fundus AF, retinal and RPE/choroidal tissues were carefully dissected and flat mounts prepared of each layer separately. Retinal vessels were revealed by staining of retinal flat mounts with anti-CD31 and/or phalloidin. An age-dependent increase in the number of AF granules was observed in the retinal perivascular areas (Fig. 2A,B) with greater concentration around retinal venules than arterioles (Fig. 2C). Staining of retinal flat mounts with anti-mouse Iba-1, a microglial cell marker (Ito et al., 1998), indicated that the AF granular deposits in the neuroretina were contained within Iba-1+ perivascular microglia (Fig. 2D). AF granules were not observed in amoeboid or ramified parenchymal microglia of any age (Fig. 2E,F).

image

Figure 2. Autofluorescence (AF) in flat mount retinas of different ages of mice. Retinal flat mounts were stained with fluorescein isothiocyanate (FITC)–anti-mouse CD31 (A and B) or FITC–anti-CD31 and Alexa Fluor 633 phalloidin (C) or Alexa Fluor 633 phalloidin and PE-anti-mouse Iba-1 (D–F) and observed by confocal microscopy. (A–C) AF signals were collected with 543 nm excitation and a BP560-615 filter. AF signals (yellow) are seen around retinal vessels (green). (A) Retinal flat mount from a 6-month-old mouse. (B) Retinal flat mount from a 24-month-old mouse. (C) Retinal flat mount from an 18-month-old mouse. (D–F) Retinal flat mounts from an 18-month-old mouse. AF signals were collected with 488 nm excitation and a BP505-530 filter. AF signals were detected in Iba-1+ (red) perivascular cells (arrows, AF: yellow). a, arteriole; v, venule.

Download figure to PowerPoint

Confocal microscopy of RPE–choroidal flat mounts revealed a population of cells that were high-AF positive, and the number of these high-AF cells increased with age (Fig. 3A,B). Surprisingly, the majority of high-AF+ cells were morphologically dendriform but not RPE-like (Fig. 3A,B). Phenotype study indicated that these dendriform cells were Iba-1+ microglia (Fig. 3C). Weak AF (low-AF+) signals were observed in almost all RPE cells in mice older than 12 months (Fig. 3A,B) but rarely in younger mice. Occasionally, a few RPE cells were also detected with high-AF+ granules within their cytoplasm (Fig. 3D). The high-AF+ RPE cells can be differentiated from subretinal microglial cells by their unique morphology as well as their anatomic locations (Fig. 3D,E). In all age groups, the AF subretinal microglia outnumbered the AF RPE cells (Fig. 3F). Statistical analyses revealed a strong correlation between the age of the animal and the number of high-AF+ microglial cells or the number of high-AF RPE+ cells (Pearson correlation test, r2 = 0.89; P = 0.02 for microglia, and r2 = 0.94; P = 0.015 for RPE), suggesting that the presence and accumulation of high AF+ cells are age dependent.

image

Figure 3. Autofluorescence (AF) in subretinal microglia and retinal pigment epithelial (RPE) cells. (A and B) RPE/Choroidal flat mounts from a 12-month- and an 18-month-old mouse were stained with Alexa Fluor 633 phalloidin (blue) and observed by confocal microscopy. AF signals were collected using both 488 nm and 543 nm excitations (yellow). (C) RPE/Choroidal flat mount from a 12-month-old mouse was stained with Alexa Fluor 633 phalloidin and PE–anti-Iba-1 (red). AF signals were collected using 488 nm excitation. AF+ particles are shown as yellow. (D and E) RPE/Choroidal flat mounts from 18-month-old mice were stained with Alexa Fluor 633 phalloidin (blue) and PI (red). AF signals were collected using 488 nm excitation (green or yellow when overlaid with red). In E, a starfish-like microglia was seen on top of RPE cells in the Z-stack images (star, RPE nucleus; arrow, microglial nucleus). (F) Quantitative data of the number of AF+ microglia and AF+ RPE cells from different ages of mice. #, no AF+ RPE cell was detected in that age group. *P < 0.05; n = 6.

Download figure to PowerPoint

Electron microscopy revealed microglia adjacent to the anterior surface of RPE cells (Fig. 4A) or within the photoreceptor outer segment (POS) matrix of the subretinal space (Fig. 4B). Microglial cells were distinguished from RPE cells by their less dense cytoplasm, extensive dendriform processes, distinctly different nuclear morphology and location. Many lipofuscin granules were seen within the microglial cell cytoplasm. Interestingly, melanosomes were also detected in those cells, suggesting that subretinal microglia may receive melanin particles from RPE cells in a manner similar to melanosome transfer between stromal melanocytes and other cell types (Hirobe, 1995; Aspengren et al., 2006; Futter, 2006). Occasionally, POS disks were seen inside the microglial cell body (Fig. 4B, insert).

image

Figure 4. Transmission electron microscopy (TEM) of subretinal microglia. Sample from an 18-month-old mouse was processed for TEM. (A) A microglia adjacent to retinal pigment epithelial (RPE) cell. (B) A microglia within the photoreceptor outer segment (POS) matrix. Lipofuscin granules (arrows) are seen in the cytoplasm of microglia and RPE cells. A phagocytosed POS disk is seen in the microglia (arrowhead in B). BM, Bruch's membrane; MG, microglia.

Download figure to PowerPoint

Increased microglia in subretinal space with age

The fact that in aged mice, the majority of high-AF+ cells in the RPE–choroidal flat mounts are microglia leads us to believe that there may be an age-dependent migration of retinal microglia to subretinal space. Staining of RPE–choroidal flat mounts from different ages of mice with Iba-1 antibody revealed an age-dependent increase in the number of Iba-1+ cells on the RPE surface (Fig. 5) (r2 = 0.86; P = 0.012; Pearson correlation test). In 3-month-old mice, an average of 12.1 ± 3.8 cells µm−2 was detected (Fig. 5A,E), whereas by 18 months, this reached 79.9 ± 26.34 cells mm−2 (Fig. 5E). In mice younger than 12 months old, majority of the Iba-1+ microglia were ramified dendriform cells (Fig. 5A,B), whereas in mice older than 18 months many Iba-1+ cells had round cell bodies and short, stubby dendrites (Fig. 5C,D). It is possible that some of the microglia on the RPE surface may detach from the tissue during sample preparation (e.g. during dissection, staining and wash). As all samples were processed using the same protocol, we believe the increased numbers of subretinal microglia reflect the true changes in aged mouse retina. However, the actual number may be somewhat higher than what we observed in this study.

image

Figure 5. Subretinal microglia in different ages of mice. Retinal pigment epithelial/Choroidal flat mounts from a 3-month- (A), 6-month- (B), 12-month- (C) and 18-month-old mice were stained with PE–anti-Iba-1 and observed by confocal microscopy. (E) Quantitative data showing the number of Iba-1+ microglia in different ages of mice. *P < 0.05; n = 6.

Download figure to PowerPoint

Dual staining of RPE–choroidal flat mounts with Iba-1 and other macrophage markers indicated that all Iba-1+ in the subretinal space co-expressed high levels of CD68 (Fig. 6A), low but detectable levels of F4/80 (Fig. 6B), no MHC-II (Fig. 6C), no CD11c (Fig. 6D) and no mannose receptor (CD206, Fig. 6E) antigens.

image

Figure 6. Phenotype of subretinal microglia. Retinal pigment epithelial/Choroidal flat mounts from 12- to 18-month-old mice were stained with fluorescein isothiocyanate–anti-Iba-1 and anti-CD68 (A), F4/80 (B), MHC-II (C), CD11c (D), or CD206 (E) and examined by confocal microscopy. The Iba-1+ subretinal microglia were CD68high (A), F4/80low (B), MHC-IIlow/– (C), CD11c (D) and CD206 (E).

Download figure to PowerPoint

Age-dependent accumulation of lipofuscin in subretinal microglia

Having characterized the phenotype of subretinal microglia, we then further investigated changes in microglial lipofuscin content with age in these cells. Lipofuscin granules were observed in subretinal microglia as early as 6 months old (Fig. 7A). However, at this time, only a few lipofuscin granules existed in each microglia (Fig. 7A). With age, the number of lipofuscin granules per cell increased. In 24-month-old mice, many intracellular lipofuscin granules merged together and occupied most of the available cytoplasmic spaces (Fig. 7B). Cells with large numbers of lipofuscin granules usually displayed a large cell body and short dendrites (Fig. 7B) compared to cells with no or only few lipofuscin granules (Fig. 7A). Quantitative analysis of the lipofuscin volume against the total cell volume in Iba-1+ cells (Fig. 7C–F) revealed an age-dependent increase in the average volume of lipofuscin granules per cell (Fig. 7F) (r2 = 0.84; P = 0.04; Pearson correlation test).

image

Figure 7. Lipofuscin volume in Iba-1+ subretinal microglia of different ages of mice. Retinal pigment epithelial/Choroidal flat mounts from a 6-month- (A), a 24-month- (B) and a 12-month-old mouse (C) were stained with anti-Iba-1 (red) and Alexa Fluor 633 phalloidin (blue). Autofluorescence (AF) signals were collected using 488 nm excitation and appeared in yellow because of Iba-1 (red) co-localization. Z-stack images were reconstructed, and the volumes of Iba-1+ signal (D) and AF+ signal (E) were quantified using ImagePro Plus system. (F) Quantitative data of lipofuscin volume in Iba-1+ cells. *P < 0.05; **P < 0.01; n = 6.

Download figure to PowerPoint

Fluorescence emission fingerprints of lipofuscin in subretinal microglia and RPE cells

Using the lambda mode of the LSM510 META confocal microscope (Carl Zeiss, Hertfordshire, UK), the fluorescent emission fingerprints of AF deposits can be identified. To determine whether the lipofuscin granules in subretinal microglia have the same AF properties as those in RPE cells, AF emission spectra from each cell type were compared. Lipofuscin granules from both types of cell could be excited by a wide range of wavelengths (Fig. 8). Importantly, their emission fingerprints at all laser excitation wavelengths tested were similar, suggesting that lipofuscin granules in both types of cells may have similar chemical components. Examination of lipofuscin granules in perivascular microglial cells showed similar patterns of autofluorescent emission spectra as those in subretinal microglia and RPE cells (data not shown).

image

Figure 8. Fluorescence emission fingerprints of lipofuscin in retinal pigment epithelial (RPE) cells and subretinal microglia. Unstained RPE/choroidal flat mounts were examined using the lambda mode of LSM510 META. Samples were excited with 458 nm, 477 nm, 488 nm, 514 nm, 543 nm and 633 nm, respectively, and the emission fingerprints from each excitation wavelength were collected. Lipofuscin granules from RPE cells and microglia have similar emission fingerprints in excitation laser tested. MG, microglia.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References

Intracellular lipofuscin accumulation is a hallmark of aging (Sohal & Brunk, 1989). In aged brain tissue, lipofuscin was observed in various cells such as neurons and glial cells (Riga et al., 2006). In the retina, however, RPE cells are well recognized as lipofuscin-producing cells (Wihlmark et al., 1996a,b), and RPE lipofuscin is believed to be the main source of fundus AF (Delori et al., 1995, 2001). Lipofuscin in other retinal cells has never been investigated. In this study, we demonstrated that in the normal aging mouse retina, although lipofuscin granules were observed in RPE cells, many more were seen in Iba-1+ microglial cells, in particular those in the perivascular and subretinal space. In the retina, there are at least three subsets of microglia: parenchymal, perivascular and subretinal microglia (Dick et al., 1995; Ng & Streilein, 2001; Zhang et al., 2005; Xu et al., 2007), distinguishable by the extent of cell surface antigen expression as well as their specific anatomical tissue location. Interestingly, very little evidence of AF was observed in parenchymal microglia, suggesting that: (i) subretinal microglia are involved in extensive photoreceptor debris removal as is recognized for RPE cells; and (ii) perivascular macrophages have a similar scavenging role perhaps relating to removal of lipoperoxide-modified membranous material from the inner retina. It is also possible that some of the perivascular high-AF+ microglia migrated from the subretinal space to the perivascular zone as has been suggested for perivascular cells seen in other retinal degenerations such as retinitis pigmentosa (clinically seen as ‘bone corpuscles’). In previous studies, these cells have been assumed to be migrating RPE cells, but it is also possible that they are melanosome-filled migratory microglia. Further investigation of this possibility using immunophenotyping with dual and triple staining techniques is required.

How lipofuscin is formed in retinal microglia is not known. In most post-mitotic cells, it is believed that lipofuscin originates from autophagocytosed cellular components that have become oxidized outside or inside the lysosomal compartment (Terman & Brunk, 1998). The perivascular zone and the subretinal space are important sites for supplying retinal nutrients and oxygen as well as removing metabolic waste materials. Retinal microglia and choroidal macrophages are important cells responsible for the removal of retinal waste materials. Microglia in those areas may be more phagocytically active, and thus become ‘overloaded’ with waste materials to a greater extent than microglia at other retinal sites. With age, these cells may become less capable of digesting phagocytosed waste material, which finally become evident as increased lipofuscin deposits. Our results suggest that microglia at those two sites are either more active or are presented with a greater phagocytic load than those at other retinal sites, and may play an important role in retinal homeostasis and in age-related retinal degeneration. In addition, it is likely that they are specifically recruited to these sites, particularly the subretinal space, in response to a failing, aging RPE monolayer, and the signals involved in this process are worthy of further study.

Subretinal microglia have been observed previously in human fetal eyes (McMenamin & Loeffler, 1990) and in developing eyes of eutherian mammals and marsupials (McMenamin, 1999). In adult mice, Ng & Streilein (2001) showed that, in response to light stimulation, retinal microglia migrated to subretinal space. Recently, in a bone marrow chimeric model, we demonstrated that retinal microglia migrated from the inner retina to the POS layer in peripheral retinal regions 6 months after bone marrow transplantation in physiological conditions (Xu et al., 2007). In pathological conditions, retinal microglia have been observed to migrate to subretinal space by many groups (Chen et al., 2002; Gupta et al., 2003; Hughes et al., 2003; Zeng et al., 2005). In this study, we observed an age-dependent migration of microglia to the subretinal space. What causes the migration of microglia to the subretinal space during normal aging is not known. It is possible that during aging, oxidized unsaturated fatty acids and other lipoproteins in the POSs accumulate (Wiegand et al., 1983; Penn et al., 1987). The oxidized materials could then stimulate RPE cells to produce various cytokines/chemokines, including granulocyte–macrophage colony-stimulating factor (Planck et al., 1993; Crane et al., 1999), RANTES (Crane et al., 1998), MCP-1 (Elner et al., 1991), IL-6 (Elner et al., 1992; Holtkamp et al., 1998), IL-8 (Elner et al., 1990; Holtkamp et al., 1998), which may induce microglia migration.

An important question arising from this study is the function of these subretinal microglia in normal aging mice. Subretinal microglia express high levels of CD68, low levels of F4/80 and no mannose receptor suggesting that they are activated. However, as they do not express MHC-II antigens, they are not traditionally activated microphages. Microglial cells in the subretinal space were detected to contain POS discs (Fig. 4B) (Ng & Streilein, 2001) as well as melanosomes (Fig. 4) indicating that they have phagocytic ability. It is possible that microglia migrate to the subretinal space primarily to help RPE cells to assist in clearance of debris or even apoptotic RPE cells. However, the ingested POS disks or other debris may not be digested properly and become lipofuscin, which in turn may change the phenotype and the function of the host microglial cells. Lipofuscin is believed to have a toxic effect on its host cells (Brunk & Terman, 2002; Gray & Woulfe, 2005); they may also stimulate (‘activate’) microglia to produce inflammatory cytokines/chemokines, which may then affect neuroretinal tissue. A previous report in the quail has shown that with age, activated microglia accumulated within the photoreceptor layer, and this effect was inversely related to the number of photoreceptors (Kunert et al., 1999). Studies in human samples including AMD (Penfold et al., 1997; Gupta et al., 2003), retinitis pigmentosa and late-onset retinal degeneration (Gupta et al., 2003) have shown that activated microglia accumulate in the outer nuclear layer in regions of ongoing rod cell death. It is, however, not known whether migration of microglia to the diseased site is the cause or the consequence of photoreceptor cell death. In this study using electron microscopy technique, we did not see any apoptotic photoreceptor cells in the regions of subretinal microglia accumulation. However, some adjacent RPE cells were degenerate (Fig. 4A). Future functional studies are required to understand the possible role of lipofuscin-containing microglia in the development of age-related retinal degeneration.

A significant question in this context is whether these observations are relevant to fundus AF in humans. The AF pattern observed in normal aged mice differs from that seen in normal human subjects. This may be attributable to the anatomic differences of mouse and human eyes (no macular in mouse eyes) as well as the differences in relation to environmental conditions, since clearly the breeding and rearing conditions relating to laboratory mice and exposure to light damage, will be different from humans. Whatever the AF pattern of mouse and human beings, it is possible however, that the fundamental pathological change may be the same. As fundus AF is now widely used in the clinic for diagnosis as well as monitoring various retinal diseases, it has been assumed that fundus AF reflects the content of lipofuscin in RPE cells. However, it is recognized that other cells might generate AF signals, and the data in this study indicate that an important source of AF, at least in the aging mouse retina, is the microglial cell. Because of the similarity of AF spectra of microglial lipofuscin and RPE lipofuscin, if similar microglial changes were to exist in the human retina, then this should be taken into account when interpreting fundus AF changes in patients with retinal diseases. Pathophysiologically, it might be important to determine whether there is a relationship between accumulating microglial cells, and the various forms of AMD, particularly geographic, atrophic age-related macular degeneration.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References

Animals

C57BL/6 mice were supplied by the Medical Research Facility at Aberdeen University. Four groups of mice (six to eight mice in each group) of different ages (3, 6, 12, 18, 24 months) were used in the experiment. All mice were housed and bred in a normal experimental room and exposed to normal light at a 12-h dark–12-h light cycle. All procedures concerning animals in this study were performed according to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and under the regulations of the United Kingdom Animal License Act 1986 (UK).

In vivo imaging fundus AF in aged mice

A custom-built confocal scanning laser ophthalmoscope (Manivannan et al., 1993) modified for imaging small animals (Xu et al., 2002) was used to image fundus AF. Mice of different ages (3, 12, 18, 24 months; three mice in each group) were used for this study. Mice were anesthetized with an intramuscular injection of 0.4 mL kg−1 Hypnorm (VetaPharma Ltd, Southampton, UK) and 1 mL kg−1 diazepam (Phoenix Pharmaceuticals Ltd, Gloucester, UK) intraperitoneally. Pupils were dilated with one drop of 1% tropicamide (Chauvin Pharmaceuticals, Surrey, UK). A contact lens was then placed on the cornea to keep the eyelids open and to prevent the evaporation of tears. A 1-mm beam with a power of 400 uW argon laser (488 nm excitation) was used to sequentially scan the fundus of the animal. A barrier filter allowing light above 520 nm was used to detect the AF. Autofluorescence images are recorded as digital images of resolution 800 × 600 pixels.

Retinal, RPE–choroidal flat mounts preparation

The mice were killed by CO2 inhalation, and the eyes enucleated and fixed with 2% (v/w) paraformaldehyde in phosphate-buffered saline (PBS) (Agar Scientific Ltd, Cambridge, UK). Retinas and RPE–choroidal were dissected as previously described (Chan-Ling, 1997; Xu et al., 2003, 2005). In brief, the cornea, crystalline lens and vitreous were removed in ice-cold PBS, and the eyecups were transferred to fresh ice-cold PBS. The retina and RPE–choroid were carefully dissected. To facilitate preparation of flat mounts of the tissues, five vertical cuts were made in the retina, whereas RPE–choroidal tissue was further cut into small pieces. Retinal or RPE–choroidal tissues were further processed for immunohistochemistry as follows.

After washing, retinal and RPE–choroidal tissues were blocked and permeabilized with 5% (w/v) bovine serum albumin with 0.3% (v/v) Triton in PBS at room temperature for 1–2 h. Samples were then incubated overnight (retinal tissues) or for 2 h (RPE–choroidal tissues) at 4 °C with different combinations of anti-mouse antibodies, followed by secondary antibody incubation (4 h for retina and 1 h for RPE–choroid at room temperature). Antibodies used were biotinated rat anti-mouse CD68 (FA-11), biotinated anti-mouse CD206 (MR5D3) and biotinated anti-mouse F4/80 (A3-1) from AbD Serotec (Oxford, UK); goat anti-Iba-1 (Abcam Ltd, Cambridge, UK); fluorescein isothiocyanate (FITC) conjugated anti-CD11c (HL3), R-phycoerythrin (R-PE) conjugated anti-I-A/I-E (M5/114) from BD Biosciences (Oxford, UK); Alexa Fluor 633 phalloidin (Invitrogen Molecular Proves, Paisley, UK). Secondary antibodies used include: R-PE conjugated streptavidin (BD Biosciences) and FITC-conjugated donkey anti-goat IgG (AbD Serotec). In some samples, cell nuclei were stained with propidium iodide (Invitrogen). After thorough washing, all tissues were flat mounted in Vectashield (Vector Laboratory Ltd, Peterborough, UK) on clean glass slides. The retinal tissues were mounted with vitreous side face up, whereas the RPE–choroidal tissues were mounted with RPE side face up. All samples were examined using a confocal scanning laser microscope, LSM510 META (Carl Zeiss).

Confocal microscopy

Antibody-stained flat mount samples were examined using the LSM510 META microscope configured to the multitrack mode. AF signals were detected with either 488 nm or 543 nm wavelength or both, with a band-pass filter of BP505-530 (for 488 nm excitation) or BP 560–615 (for 543 nm excitation). Z-stack images were collected from all flat mount samples.

The RPE and microglial AF emission fingerprinting were detected in unstained RPE–choroidal flat mounts using the lambda mode LSM510 META confocal microscope. With the lambda META scanning mode, the emission signals are detected by a polychromatic 32-channel detector, which allows fast acquisition of lambda stacks. The excitation lasers 448 nm, 477 nm, 488 nm, 514 nm, 543 nm and 633 nm were used. For each laser excitation, the filter was set to allow light above the excitation wavelength to pass up to 790 nm. For all lambda acquisitions, the pinhole was set to 1 airy unit and the scanning speed was set to 6. To be able to compare the emission fingerprinting acquired from RPE cells with that from microglial cells, the detector gain, amplifier offset and amplifier gain were kept at the same levels during the acquisition.

Transmission electron microscopy

Mouse was perfused with 2.5% glutaraldehyde in PBS through the left ventricle under general anesthesia. Eyes were removed and immersed in the same fixative for at least 48 h. After thorough washing, the anterior segment, lens and vitreous were removed. The posterior segment of the mouse eye was cut into small pieces and subsequently processed for electron microscopic analysis. Ultrathin sections were stained with uranyl acetate and lead citrate, and examined with a Phillips CM 10 transmission electron microscope (Eindhoven, The Netherlands).

Image analysis

Z-stack images were reconstructed using the lsm Examiner software (Carl Zeiss). To quantify the number of cells in the RPE–choroidal flat mounts, three images were taken from six regions (five peripheral regions plus central region) of each flat mount using ×20 objective lens. Data were expressed as mean ± SEM mm−2. The volume of AF granules in microglial cells was quantified using the Image Pro Plus System (Media Cybernetics, Bethesda, MD, USA). After defining the AF positive and Iba-1 positive signals, the AF volume in Iba-1+ microglia was automatically calculated and expressed as percentage of the total cell volume. Tukey's multiple comparison test was used to compare the difference between different age groups. Correlation between the age of mice and the number of cells (subretinal AF microglia and AF RPE) or the AF volume was analyzed by the Pearson correlation test. Probability values of P < 0.05 were considered statistically significant.

Acknowledgments

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References

This project was supported by the NHS Endowment. Dr Heping Xu is a Research Council UK academic fellow supported by the Department of Trade and Industry and Office of Science and Technology.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  • Aspengren S, Hedberg D, Wallin M (2006) Studies of pigment transfer between xenopus laevis melanophores and fibroblasts in vitro and in vivo. Pigment Cell Res. 19, 136145.
  • Bellmann C, Rubin GS, Kabanarou SA, Bird AC, Fitzke FW (2003) Fundus autofluorescence imaging compared with different confocal scanning laser ophthalmoscopes. Br. J. Ophthalmol. 87, 13811386.
  • Bindewald A, Bird AC, Dandekar SS, Dolar-Szczasny J, Dreyhaupt J, Fitzke FW, Einbock W, Holz FG, Jorzik JJ, Keilhauer C, Lois N, Mlynski J, Pauleikhoff D, Staurenghi G, Wolf S (2005) Classification of fundus autofluorescence patterns in early age-related macular disease. Invest. Ophthalmol. Vis. Sci. 46, 33093314.
  • Brunk UT, Terman A (2002) Lipofuscin: mechanisms of age-related accumulation and influence on cell function. Free Radic. Biol. Med. 33, 611619.
  • Chan-Ling T (1997) Glial, vascular, and neuronal cytogenesis in whole-mounted cat retina. Microsc. Res. Tech. 36, 116.
  • Chen L, Yang P, Kijlstra A (2002) Distribution, markers, and functions of retinal microglia. Ocul. Immunol. Inflamm. 10, 2739.
  • Crane IJ, Kuppner MC, McKillop-Smith S, Knott RM, Forrester JV (1998) Cytokine regulation of RANTES production by human retinal pigment epithelial cells. Cell Immunol. 184, 3744.
  • Crane IJ, Kuppner MC, McKillop-Smith S, Wallace CA, Forrester JV (1999) Cytokine regulation of granulocyte–macrophage colony-stimulating factor (GM-CSF) production by human retinal pigment epithelial cells. Clin. Exp. Immunol. 115, 288293.
  • Delori FC, Dorey CK, Staurenghi G, Arend O, Goger DG, Weiter JJ (1995) In vivo fluorescence of the ocular fundus exhibits retinal pigment epithelium lipofuscin characteristics. Invest. Ophthalmol. Vis. Sci. 36, 718729.
  • Delori FC, Goger DG, Dorey CK (2001) Age-related accumulation and spatial distribution of lipofuscin in RPE of normal subjects. Invest. Ophthalmol. Vis. Sci. 42, 18551866.
  • Dick AD, Ford AL, Forrester JV, Sedgwick JD (1995) Flow cytometric identification of a minority population of MHC class II positive cells in the normal rat retina distinct from CD45lowCD11b/c+CD4low parenchymal microglia. Br. J. Ophthalmol. 79, 834840.
  • Dorey CK, Wu G, Ebenstein D, Garsd A, Weiter JJ (1989) Cell loss in the aging retina. Relationship to lipofuscin accumulation and macular degeneration. Invest. Ophthalmol. Vis. Sci. 30, 16911699.
  • Dowson JH (1982) Neuronal lipofuscin accumulation in ageing and Alzheimer dementia: a pathogenic mechanism? Br. J. Psychiatry 140, 142148.
  • Elner VM, Strieter RM, Elner SG, Baggiolini M, Lindley I, Kunkel SL (1990) Neutrophil chemotactic factor (IL-8) gene expression by cytokine-treated retinal pigment epithelial cells. Am. J. Pathol. 136, 745750.
  • Elner SG, Strieter RM, Elner VM, Rollins BJ, Del Monte MA, Kunkel SL (1991) Monocyte chemotactic protein gene expression by cytokine-treated human retinal pigment epithelial cells. Lab. Invest. 64, 819825.
  • Elner VM, Scales W, Elner SG, Danforth J, Kunkel SL, Strieter RM (1992) Interleukin-6 (IL-6) gene expression and secretion by cytokine-stimulated human retinal pigment epithelial cells. Exp. Eye Res. 54, 361368.
  • Futter CE (2006) The molecular regulation of organelle transport in mammalian retinal pigment epithelial cells. Pigment Cell Res. 19, 104111.
  • Gray DA, Woulfe J (2005) Lipofuscin and aging: a matter of toxic waste. Sci. Aging Knowledge Environ. 2005, re1.
  • Gupta N, Brown KE, Milam AH (2003) Activated microglia in human retinitis pigmentosa, late-onset retinal degeneration, and age-related macular degeneration. Exp. Eye Res. 76, 463471.
  • Hirobe T (1995) Structure and function of melanocytes: microscopic morphology and cell biology of mouse melanocytes in the epidermis and hair follicle. Histol. Histopathol. 10, 223237.
  • Holtkamp GM, Van Rossem M, De Vos AF, Willekens B, Peek R, Kijlstra A (1998) Polarized secretion of IL-6 and IL-8 by human retinal pigment epithelial cells. Clin. Exp. Immunol. 112, 3443.
  • Hughes EH, Schlichtenbrede FC, Murphy CC, Sarra GM, Luthert PJ, Ali RR, Dick AD (2003) Generation of activated sialoadhesin-positive microglia during retinal degeneration. Invest. Ophthalmol. Vis. Sci. 44, 22292234.
  • Ito D, Imai Y, Ohsawa K, Nakajima K, Fukuuchi Y, Kohsaka S (1998) Microglia-specific localisation of a novel calcium binding protein, Iba1. Brain Res. Mol. Brain Res. 57, 19.
  • Kennedy CJ, Rakoczy PE, Constable IJ (1995) Lipofuscin of the retinal pigment epithelium: a review. Eye 9, 763771.
  • Kunert KS, Fitzgerald ME, Thomson L, Dorey CK (1999) Microglia increase as photoreceptors decrease in the aging avian retina. Curr. Eye Res. 18, 440447.
  • Lois N, Halfyard AS, Bird AC, Fitzke FW (2000) Quantitative evaluation of fundus autofluorescence imaged ‘in vivo’ in eyes with retinal disease. Br. J. Ophthalmol. 84, 741745.
  • Lois N, Owens SL, Coco R, Hopkins J, Fitzke FW, Bird AC (2002) Fundus autofluorescence in patients with age-related macular degeneration and high risk of visual loss. Am. J. Ophthalmol. 133, 341349.
  • Lois N, Halfyard AS, Bird AC, Holder GE, Fitzke FW (2004) Fundus autofluorescence in Stargardt macular dystrophy–fundus flavimaculatus. Am. J. Ophthalmol. 138, 5563.
  • Manivannan A, Sharp PF, Phillips RP, Forrester JV (1993) Digital fundus imaging using a scanning laser ophthalmoscope. Physiol. Meas. 14, 4356.
  • McMenamin PG (1999) Subretinal macrophages in the developing eye of eutherian mammals and marsupials. Anat. Embryol. (Berl.) 200, 551558.
  • McMenamin PG, Loeffler KU (1990) Cells resembling intraventricular macrophages are present in the subretinal space of human foetal eyes. Anat. Rec. 227, 245253.
  • Ng TF, Streilein JW (2001) Light-induced migration of retinal microglia into the subretinal space. Invest. Ophthalmol. Vis. Sci. 42, 33013310.
  • Penfold PL, Liew SC, Madigan MC, Provis JM (1997) Modulation of major histocompatibility complex class II expression in retinas with age-related macular degeneration. Invest. Ophthalmol. Vis. Sci. 38, 21252133.
  • Penn JS, Naash MI, Anderson RE (1987) Effect of light history on retinal antioxidants and light damage susceptibility in the rat. Exp. Eye Res. 44, 779788.
  • Planck SR, Huang XN, Robertson JE, Rosenbaum JT (1993) Retinal pigment epithelial cells produce interleukin-1 beta and granulocyte–macrophage colony-stimulating factor in response to interleukin-1 alpha. Curr. Eye Res. 12, 205212.
  • Riga D, Riga S, Halalau F, Schneider F (2006) Brain lipopigment accumulation in normal and pathological aging. Ann. N.Y. Acad. Sci. 1067, 158163.
  • Von Ruckmann A, Fitzke FW, Bird AC (1995) Distribution of fundus autofluorescence with a scanning laser ophthalmoscope. Br. J. Ophthalmol. 79, 407412.
  • Scholl HP, Chong NH, Robson AG, Holder GE, Moore AT, Bird AC (2004) Fundus autofluorescence in patients with leber congenital amaurosis. Invest. Ophthalmol. Vis. Sci. 45, 27472752.
  • Sohal RS, Brunk UT (1989) Lipofuscin as an indicator of oxidative stress and aging. Adv. Exp. Med. Biol. 266, 1726.
  • Sparrow JR, Boulton M (2005) RPE lipofuscin and its role in retinal pathobiology. Exp. Eye Res. 80, 595606.
  • Strehler BL, Mark DD, Mildvan AS (1959) GEE MV: rate and magnitude of age pigment accumulation in the human myocardium. J. Gerontol. 14, 430439.
  • Terman A, Brunk UT (1998) Lipofuscin: mechanisms of formation and increase with age. APMIS 106, 265276.
  • Wiegand RD, Giusto NM, Rapp LM, Anderson RE (1983) Evidence for rod outer segment lipid peroxidation following constant illumination of the rat retina. Invest. Ophthalmol. Vis. Sci. 24, 14331435.
  • Wihlmark U, Wrigstad A, Roberg K, Brunk UT, Nilsson SE (1996a) Lipofuscin formation in cultured retinal pigment epithelial cells exposed to photoreceptor outer segment material under different oxygen concentrations. APMIS 104, 265271.
  • Wihlmark U, Wrigstad A, Roberg K, Brunk UT, Nilsson SE (1996b) Formation of lipofuscin in cultured retinal pigment epithelial cells exposed to pre-oxidized photoreceptor outer segments. APMIS 104, 272279.
  • Wolf G (2003) Lipofuscin and macular degeneration. Nutr. Rev. 61, 342346.
  • Woon WH, Fitzke FW, Bird AC, Marshall J (1992) Confocal imaging of the fundus using a scanning laser ophthalmoscope. Br. J. Ophthalmol. 76, 470474.
  • Xu H, Manivannan A, Goatman KA, Liversidge J, Sharp PF, Forrester JV, Crane IJ (2002) Improved leukocyte tracking in mouse retinal and choroidal circulation. Exp. Eye Res. 74, 403410.
  • Xu H, Forrester JV, Liversidge J, Crane IJ (2003) Leukocyte trafficking in experimental autoimmune uveitis: breakdown of blood–retinal barrier and upregulation of cellular adhesion molecules. Invest. Ophthalmol. Vis. Sci. 44, 226234.
  • Xu H, Dawson R, Crane IJ, Liversidge J (2005) Leukocyte diapedesis in vivo induces transient loss of tight junction protein at the blood–retina barrier. Invest. Ophthalmol. Vis. Sci. 46, 24872494.
  • Xu H, Chen M, Mayer EJ, Forrester JV, Dick AD (2007) Turnover of resident retinal microglia in the normal adult mouse. Glia 55, 11891198.
  • Zeng HY, Zhu XA, Zhang C, Yang LP, Wu LM, Tso MO (2005) Identification of sequential events and factors associated with microglial activation, migration, and cytotoxicity in retinal degeneration in rd mice. Invest. Ophthalmol. Vis. Sci. 46, 29922999.
  • Zhang C, Shen JK, Lam TT, Zeng HY, Chiang SK, Yang F, Tso MO (2005) Activation of microglia and chemokines in light-induced retinal degeneration. Mol. Vis. 11, 887895.