An improved method for the isolation and culture of retinal pigment epithelial cells from adult rats.

PURPOSE: Since adult rats are used in pre-clinical studies, and due to the necessity of investigating the side-effects of drugs on RPE cells in vitro, there is a great need for primary RPE cells from these animals. The aim of this study was to develop a reproducible and quantifiable method of isolation, culture, and maintenance of adult rat RPE cells. Moreover, potential differences between RPE cells from albino versus pigmented rats were also investigated. METHODS: A total of 180 pigmented rats and 340 albino rats aged 6-14 weeks were used. RPE cells were isolated and cultured for several weeks by using three different methods: 1) growing directly on flat mounts, 2) after enzymatic isolation, and 3) after they spontaneously detached from the flat mounts and continued to grow on the plastic. Yield, cell survival, and morphological characteristics were investigated using light and electron microscopy as well as immunohistochemistry. RESULTS: After 0 weeks, the yield of the first method was 30,000 cells/eye; after 2 weeks 18,000 cells/eye; and after 4 weeks 11,000 cells/eye. The yield of RPE cells was very low after enzymatic isolation in method 2 (0 weeks, 13.000 cells/eye; 2 weeks, 30,000 cells/eye; 4 weeks 38,000 cells/eye), whereas it was higher when the RPE cells spontaneously detached from the flat mounts and then continued to grow on the plastic in method three. (0 weeks, 30,000 cells/eye; 2 weeks, 314,000 cells/eye; 4 weeks, 659,000 cells/eye). The second method often showed contamination with fibroblasts, whereas the two other methods showed pure RPE cultures. The RPE cells were able to proliferate when using the second and the third method, but not when they were cultivated directly on the flat mounts (first method). CONCLUSION: The qualitative and quantitative best method for isolating adult rat RPE cells is the culture of RPE cells which spontaneously detach from flat mounts. No differences were observed between albino and pigmented RPE cells.

A new kind of labyrinth-like capillary is responsible for leakage from human choroidal neovascular endothelium, as investigated by high-resolution electron microscopy.

PURPOSE: This study reports the clinicopathologic findings of leaky sites in pathological vessels after submacular removal of choroidal neovascular membranes (CNV). As the site that causes fluid exudation from neovascular vessels is unknown, specific attention was focused on the formation of fenestrations, cellular junctions, and morphologic alteration which can cause endothelial leakage. METHODS: Choroidal neovascular membranes of 15 patients who underwent submacular surgery for CNV were investigated. Five patients received bevacizumab treatment before surgery, and another five received photodynamic therapy before surgery. The remaining five did not receive any other treatment before surgery. All membranes were embedded for transmission electron microscopy. CNVs were analyzed for pathological cell-to-cell connections, fenestrations, or other pathological conditions which can cause leakage of plasma. RESULTS: The morphology of the newly formed blood channels was very variable, and in principle was not different in treated and untreated patients. The sources of leakage in neovascular choroidal vessels were caused by insufficient endothelial cell connections and by capillaries with microvillar projections into the vessel lumen which blocked cellular perfusion but still allowed the flow of plasma. Fenestrations were only infrequently observed. CONCLUSIONS: A newly discovered type of pathological capillary, called a labyrinth capillary, is very likely responsible for the permanent leakage of fluid. Due to the small vessel lumen, thrombocytes cannot enter these capillaries to close the leakages. Fenestrations did not appear to play a significant role in vascular leakiness.

Formation of immune complexes and thrombotic microangiopathy after intravitreal injection of bevacizumab in the primate eye.

BACKGROUND: In this study, the effect of intravitreal injection of bevacizumab on choroidal blood vessels was examined in primate eyes. METHODS: Four Cynomolgus monkeys received an intravitreal injection of 1.25 mg bevacizumab. The eyes were enucleated on days 1, 4, 7 and 14. For each animal, one eye was embedded in paraffin whereas the other eye was embedded for electron microscopy. Seven untreated or PBS (phosphate buffered saline)-injected monkeys were used as controls. RESULTS: Thrombotic microangiopathy was found in the choriocapillaris and choroidal vessels of all eight injected eyes. Acute microangiopathy was characterized ultrastructurally as swelling of the endothelium, loss of fenestrations and complete collapse of the capillaries, and was commonly observed in bevacizumab-treated eyes. Quantitative analysis showed reduction of the lumina of the choriocapillaris in the eyes of three of the monkeys. Bevacizumab was frequently localized inside the blood vessels, often filling the entire breadth of the vessels, and formed clusters with blood cells. Death of photoreceptors occurred in two monkeys. CONCLUSIONS: This study indicate that intravitreal injection of bevacizumab in monkeys induces activation of platelets, degranulation of thrombocytes and neutrophils, formation of immune complexes, thrombotic microangiopathy and alteration of the blood flow in choroidal vessels.

RGMA and neogenin protein expression are influenced by lens injury following optic nerve crush in the rat retina.

BACKGROUND: The death and the failure of neurons to regenerate their axons after lesion of the central nervous system in mammals, as in the case of spinal cord injury and optic nerve trauma, remain a challenge. In this study, we focused on the repulsive guidance molecule A (RGMA) and its receptor neogenin. Since it was reported that RGMA+ cells accumulate in lesioned areas after spinal cord injury, brain trauma, and optic nerve crush, and curiously, anti-apoptotic effects of RGMA were also described, we investigated the role of RGMA and neogenin in the retina after optic nerve crush (ONC). METHODS: We evaluated the spatial and temporal protein pattern of RGMA and neogenin in the rat retina without (non-regenerating model) or with (regenerating model) lens injury (LI). We investigated the presence of RGMA, neogenin and other proteins at up to nine time points (6 h-20 days post-surgery) by performing immunohistochemistry and Western blots. RESULTS: Independent of the treatment, RGMA protein was present in the nuclear layers (NLs), plexiform layers (PLs), nerve fiber layer (NFL), and in retinal ganglion cells (RGCs) of the rat retina. RGC and nerve fibers were always RGMA+. Further RGMA+ cells in the retina were blood vessel endothelial cells, astrocytes, Müller cells, and some microglial cells. The RGMA pattern for the specific retinal cells resembled those of previously published data. The neogenin pattern was congruent to the RGMA pattern. Western blots of retinal tissue showed further RGMA+ products only in LI animals. Furthermore, a higher amount of RGMA was found in the retinae of ONC + LI rats compared to ONC rats. CONCLUSIONS: Although a difference in the localization of RGMA is not obvious, the difference in the amount of RGMA is striking, the higher amount of RGMA in the retinae of ONC + LI rats compared to ONC rats indicates a role for RGMA during degeneration/regeneration processes. Our results are consistent with several reported neuroprotective effects of RGMA. Our new data showing the upregulation of RGMA after ONC in our regenerating model (plus LI) confirm these findings conducted in different settings.

Different spatial and temporal protein expressions of repulsive guidance molecule a and neogenin in the rat optic nerve after optic nerve crush with and without lens injury.

The failure of lesioned mammalian CNS neurons to regenerate their axons remains a challenge. Evidence is emerging that repulsive proteins contribute to this failure. The repulsive guidance molecule A (RGMA) induces growth cone collapse in vitro, accumulates in the scar after spinal cord injury, and is up-regulated in glaucoma. In this study, we evaluated the spatial and temporal localization pattern of RGMA and its receptor neogenin in the optic nerve after optic nerve crush (ONC) without or with lens injury (LI) at up to nine time points (6 hr to 20 days) postsurgery by performing immunohistochemistry and Western blots. We found RGMA at the crush site (CS) and in the developing scar of ONC rats at every time point investigated, whereas it was absent in the CS of ONC + LI rats. Independent of the model, many cells were RGMA(+) in the ON: nerve fibers, blood vessels, astrocytes, oligodendrocytes, some microglia, some macrophages, and the sheath of the ON. Western blots showed a significantly lowered amount of RGMA in ONC + LI animals at 2, 4, and 6 days after crush compared with ONC animals. Furthermore, LI in sham-operated animals showed an increase of RGMA in six of eight time points compared with the sham-operated animals. Moreover, the effects of LI on the morphology of the ON were characterized at a level of detail never reported before. Our results show that RGMA is present and might contribute to the inhibitory environment in the ON, especially in and around the CS after ONC.

Loss of retinal function in aged DBA/2J mice - New insights into retinal neurodegeneration.

The DBA/2J mouse is a common animal model of glaucoma. The intraocular pressure increases with age, and retinal ganglion cells (RGC) degenerate, usually starting at an age of approximately six months. In this study, we used two-year-old DBA/2J mice presuming an end-point of RGC degeneration. We investigated visual function in these animals using electroretinography (ERG) and visual evoked potentials (VEP), and we checked the number of remaining RGC by retrograde staining. Almost no RGC were left in the retina, and VEP were hardly recordable. Surprisingly, also ERG amplitudes of scotopic a-waves and b-waves, photopic b-waves and oscillatory potentials were decreased significantly by approximately 40% compared to amplitudes measured in age-matched C57BL/6J mice. The latencies were not changed in DBA/2J mice compared to C57BL/6J mice, and so were the ratios between amplitudes of a-waves, b-waves and oscillatory potentials. Our results indicate that, in addition to degeneration of RGC, also photoreceptors are affected by pathological processes in the eye caused by the mutations present in DBA/2J mice.

Ultrastructural analysis of the pigment dispersion syndrome in DBA/2J mice.

PURPOSE: To characterise ocular pigment abnormalities associated with iris atrophy in DBA/2J mice as a model for human pigment dispersion syndrome. METHODS: Immunohistochemistry, electron and light microscopy were performed to examine the eyes of DBA/2J mice ranging in age from 2.5 to 18 months old. The focus of our study was the description of the ultrastructural modifications in the irides of DBA/2J mice. RESULTS: The DBA/2J mice presented modifications in the melanosomes in all the pigmented parts of the eye, including the retinal pigment epithelial cells and choroidal melanocytes of the ciliary pigment epithelium. The extracellular matrix of the iris stroma disappeared with ageing. Pigmented cells detached from the iris and migrated into the trabecular meshwork exclusively on the anterior iris surface. These cells were identified as macrophages by immunohistochemistry and electron microscopy. There was no evidence that melanocytes or iris pigment epithelial cells migrated into the trabecular meshwork, but they became more and more depigmented. The aqueous outflow was blocked by pigment-laden cells, but not by cellular debris or melanosomes. No substantial amount of extracellular melanosomes was observed. CONCLUSION: The morphology of melanosomes is aberrant in all pigment cells in the eyes of DBA/2J mice. We conclude that the disease process begins with the transfer of both immature melanosomes from the iris pigment epithelium (IPE) and melanocytes to macrophages, which subsequently migrate into the trabecular meshwork. Accumulating macrophages cause a blockade of the chamber angle. As the disease progresses, the IPE, melanocytes and iris stroma, including blood vessels, disappear, leading to iris atrophy. It is speculated that the loss of these pigment cells is partly caused by reduction of the iris stroma.

Purkinje cell survival in organotypic cultures: implication of Rho and its downstream effector ROCK.

Organotypic cultures of postnatal day 1 (P1) to P7 mouse cerebella are well-established models for studying cell survival. In the present work, we investigate the involvement of the Rho/ROCK intracellular pathway in Purkinje cell survival by using organotypic cultures of P3 Swiss mice. Specific inhibitors of Rho or ROCK were applied at different concentrations to the slice cultures, which were maintained for 5 days in vitro. We show that the bacterial exoenzyme C3 transferase, a specific inhibitor of the small GTPase Rho, increases Purkinje cell survival. There is a 4.5- and 2.5-fold increase in Purkinje cell survival when C3 intracellular uptake is promoted either by the PEP-1 peptide or by the C2IN carrier protein, respectively, and not with the commonly used TAT peptide. Moreover, treatment with Y27632 and H-1152, two specific inhibitors of the Rho kinase ROCK, also strongly reduces apoptotic cell death and results in 6.5- and 8.5-fold increases in cell survival, respectively. In immunohistochemical analysis, we also show that H-1152 did not change either glial fibrillary acidic protein or isolectin-B4 staining, indicating that this compound did not alter the cellular composition in our cultures. Thus, our data demonstrate that inhibition of Rho and its downstream effector ROCK may be used to enhance cell survival in neurodegenerative diseases.

A reproducible and quantifiable model of choroidal neovascularization induced by VEGF A165 after subretinal adenoviral gene transfer in the rabbit.

PURPOSE: To determine the effects of the vascular endothelial growth factor (VEGF)-A(165) delivered using a high capacity adenoviral vector (HC Ad.VEGF-A) on vascular growth and pathological changes in the rabbit eye. To combine different detection methods of VEGF-A(165) overexpression-induced neovascularization in the rabbit. METHODS: HC Ad.VEGF-A(165) was constructed and injected at 5 x 10(6) infectious units (iu) into the subretinal space of rabbit eyes. Two and four weeks postinjection, the development of neovascularization and the expression of HC Ad-transduced VEGF-A(165) protein were followed up in vivo by scanning laser ophthalmoscopy, fluorescein and indocyanine green angiographies and ex vivo by electron microscopy and immunohistochemistry RESULTS: We observed a choroidal neovascularization (CNV) with leakage in 83% of the rabbit eyes. Our findings present clear indications that there is a significant effect on the endothelial cells of the choriocapillaris after subretinal transduction of the retinal pigment epithelium (RPE) with VEGF-A(165) vector. The choroidal endothelial cells were activated, adherent junctions opened, and the fenestration was minimized, while the extracellular matrix localized between the RPE and the endothelium of the choriocapillaris was enlarged toward the lumen of the vessels, inducing a deep invagination of the endothelial cells into the vessel lumen. They also proliferated and formed pathological vessels in the subretinal space. Moreover,there was an increased expression of basic fibroblast growth factor and VEGF-A accompanied by macrophage stimulation, retinal edema, and photoreceptor loss. CONCLUSIONS: This is the first model of VEGF-induced CNV in the rabbit in which the pathological events following overexpression of VEGF by RPE cells have been described in detail. Many of the features of our experimental CNV resemble those observed clinically in patients having wet age-related macular degeneration.

UV-A induced oxidative stress is more prominent in naturally pigmented aged human RPE cells compared to non-pigmented human RPE cells independent of zinc treatment.

To investigate the effects of zinc supplementation on human amelanotic (ARPE-19) and native pigmented retinal pigment epithelial cells (hRPE) under normal light conditions and after ultraviolet A light exposure. hRPE cells, containing both melanin and lipofuscin granules, were prepared from human donor eyes of 60-70 year old patients. Cells of the amelanotic ARPE-19 cell line and pigmented hRPE cells were treated with zinc chloride and subjected to oxidative stress by UV-A irradiation. Intracellular H(2)O(2) formation was measured using a fluorescence oxidation assay. Additionally, apoptosis and viability assays were performed. Control cells were treated identically except for irradiation and zinc supplementation. Under normal light conditions, zinc treated hRPE cells produced less H(2)O(2) than unsupplemented hRPE cells. Viability and apoptosis events did not change. After UV-A irradiation, ARPE and hRPE cells were greatly impaired in all tests performed compared to the non-irradiated controls. No differences were found after zinc supplementation. hRPE cells showed a higher apoptosis and mortality rate than non-pigmented cells when stressed by UV-A light. ARPE cells never showed any zinc related effects. In contrast, without irradiation, zinc supplementation reduced H(2)O(2) production in pigmented hRPE cells slightly. We did not find any zinc effect in irradiated hRPE cells. After UV light exposure, pigmented cells showed a higher apoptosis and mortality than cells lacking any pigmentation. We conclude that cells with pigmentation consisting of melanin and lipofuscin granules have more prooxidative than antioxidative capacity when stressed by UV light exposure compared to cells lacking any pigmentation.

Gene expression of the repulsive guidance molecules/neogenin in the developing and mature mouse visual system: C57BL/6J vs. the glaucoma model DBA/2J.

We used in-situ hybridization to analyze the expression patterns of three known members (a, b and c) of the RGM ("repulsive guidance molecule") gene family and of the RGMa receptor neogenin in a glaucoma mouse model (DBA/2J strain) and the C57BL/6J strain, which served as a control. In order to understand the role of the RGMs and neogenin in glaucoma, we characterized their expression patterns in the developing and mature mouse retina and in the optic nerve. In all investigated stages from post-natal day (P) 0 to 15 months (M) RGMa, RGMb and neogenin expression was detected in the ganglion cell layer (GCL). From P10 to 15M, we found RGMa, RGMb and neogenin expression in the inner nuclear layer (INL) and the outer nuclear layer (ONL). In P10- and older mice, the expression patterns of RGMa and its receptor neogenin were similar, while that of RGMb differed from both. As expected, no specific retinal expression of RGMc was detected in any of the age groups investigated. C57BL/6J mice and DBA/2J mice displayed no differences in the expression pattern of RGMa, RGMb, RGMc and neogenin in the developing retina (gestational age 14.5 days (E14.5), P0 & P10). Interestingly, we found a higher expression of RGMa, RGMb and neogenin in the retinas of all glaucoma-affected mice than in the age-matched control strain. Furthermore, we detected a higher RGMa and RGMb expression in the optic nerves of glaucoma-affected DBA/2J-mice older than 11M than in C57BL/6J mice of the same age.

Penetration of bevacizumab through the retina after intravitreal injection in the monkey.

PURPOSE: The penetration of intravitreally injected bevacizumab in its commercial formulation (Avastin; Roche, Grenzach, Germany) through the retina was studied, to determine whether a full-length antibody would be able to penetrate the retina as easily as an antibody fragment. METHODS: Six cynomolgus monkeys (Macaca fascicularis) were used in this study. Two compositions of intravitreal injection into the right eyes were performed: one with commercial Avastin (group 1, four animals) and the other one with commercial Avastin labeled with 125I (group 2, one animal). The animals in group 1 were killed 1, 4, 7, or 14 days after the injection for subsequent histologic analysis of the eyes by immunohistochemistry, and the animal in group 2 was killed 7 days after injection for autoradiography and electron microscopy. Funduscopy was performed before the injection and at several time points thereafter. Moreover, blood samples were collected at different time points from the group-2 animal. The sixth animal remained untreated and served as the control. RESULTS: No pathologic changes were obvious in the funduscopic images within the time of the experiment. Bevacizumab immunoreactivity was found in the choroid and the inner layers of the retina as early as 1 day after the injection and spread to the outer layers and the choroid within the following days, in particular to photoreceptors and blood vessels. Avastin labeled with 125I showed radioactivity in blood serum 1 day after the intravitreal injection and remained relatively stable until day 7. CONCLUSIONS: The results clearly show that the bevacizumab molecule can penetrate the retina and is also transported into the retinal pigment epithelium, the choroid and, in particular, into photoreceptor outer segments after intravitreal injection of Avastin. Active transport mechanisms seem to be involved.

Ultrastructural findings in the primate eye after intravitreal injection of bevacizumab.

PURPOSE: To examine the ultrastructural effect of intravitreal bevacizumab on primate eyes with particular focus set on the choriocapillaris and to examine the influence of vascular endothelial growth factor (VEGF) inhibition on endothelial cell fenestration. DESIGN: Animal study. METHODS: Four Cynomolgus monkeys received an intravitreal injection of 1.25 mg bevacizumab. The eyes were enucleated and prepared for light and electron microscopy on days one, four, seven, and 14. Control eyes remained untreated. Choriocapillaris endothelial cell fenestrations were quantified. RESULTS: Choriocapillaris endothelial cell fenestrations were significantly reduced after intravitreal injection of bevacizumab. Fenestration was lowest on day four (15.9 +/- 6.7 per 25 microm) and increased again from days seven to 14, but was still significantly lower than in the control (66.2 +/- 9.5 per 25 microm). Densely packed thrombocytes and leukocytes regionally occluded the choriocapillaris lumen of treated eyes. On day one an increased number of leukocytes filled in the choriocapillaris lumen. Photoreceptors were damaged in two of 40 light microscopic sections. On days one to seven, choroidal melanocytes contained giant melanosomes. None of these described features was found in controls. CONCLUSIONS: Intravitreal bevacizumab causes ultrastructural changes in the choriocapillaris of primate eyes. A significant reduction of choriocapillaris endothelial cell fenestrations is seen as early as 24 hours after injection and their number increases again after two weeks. These findings may play a role in the early clinical effect of intravitreal bevacizumab for macular edema. Because an increased risk of circulation disturbances in the choriocapillaris cannot be excluded, patients should be carefully monitored.

Antipermeability and antiproliferative effects of standard and frozen bevacizumab on choroidal endothelial cells.

BACKGROUND: Bevacizumab is an antiangiogenic compound developed to target tumour vessels. Its off-label use in ophthalmology requires in vitro testing on ocular cells. AIM: To quantify the antipermeability and antiproliferative effects of bevacizumab on cultured choroidal endothelial cells (CECs). It was examined whether deep-freezing of bevacizumab attenuates its antiangiogenic activity. METHODS: Porcine CECs were cultured in permeable insert systems. Permeability of the cell monolayers was quantified by a fluorescent isothiocyanate-dextran assay after treatment with vascular endothelial growth factor (VEGF; 20-100 ng/ml) alone and in combination with bevacizumab (0.1-1 mg/ml). Proliferation of the CECs was tested using a "wound scratch" assay. The experiments were repeated with bevacizumab after freezing at -20 degrees C for 5 days. RESULTS: Bevacizumab significantly reduced VEGF-induced permeability in a dose-dependant manner. A molar ratio of 2.6:1 of bevacizumab to VEGF was required for complete blocking of VEGF-induced rise in permeability. CEC proliferation was significantly blocked by bevacizumab (0.5 mg/ml). Thawed bevacizumab after deep freezing showed a moderate, but not statistically significant loss in activity. CONCLUSION: Bevacizumab significantly reduces VEGF-induced permeability and proliferation of CECs. Freezing and thawing of bevacizumab will affect its biological activity.

Different effects of intravitreally injected ranibizumab and aflibercept on retinal and choroidal tissues of monkey eyes.

BACKGROUND: Since there is evidence that the Fc domain of antivascular endothelial growth factor drugs may cause unexpected consequences in retinal and choroidal vessels, the effects of intravitreal ranibizumab and aflibercept on monkey eyes were investigated. METHODS: Four cynomolgus monkeys were intravitreally injected with 0.5 mg of ranibizumab and another four with 2 mg of aflibercept. Two untreated monkeys served as controls. Funduscopy, fluorescein angiography (FA), spectral-domain-optical coherence tomography (SD-OCT) and measurement of intraocular pressure (IOP) were performed. The eyes were inspected by light, fluorescence and electron microscopy. The diameter of the choriocapillaris (CC) was measured by morphometry, and the areas of the CC with free haemoglobin, CC fenestrations and endothelial thickness were quantified. RESULTS: Analysis showed ranibizumab permeated the retina via intercellular clefts, whereas aflibercept was taken up by ganglion cells, cells of the inner and outer retinal layers and the retinal pigment epithelium (RPE). Stasis and haemolysis in the choriocapillaris and choroidal vessels were more frequent after aflibercept treatment, which caused hypertrophy and death of individual RPE cells. The area of the CC was significantly reduced after both drugs compared with controls, but the reduction of the CC endothelium thickness, number of fenestrations and the areas with haemolysis were more pronounced after aflibercept. CONCLUSIONS: Ranibizumab permeated the retina through intercellular spaces, whereas aflibercept was taken up by neuronal and RPE cells. Aflibercept induced protein complex formation and more haemolysis in the choriocapillaris, leading to individual RPE cell death. The clinical significance and relation of these findings to the Fc domain or to other characteristics of aflibercept remain to be investigated.

Choriocapillaris breakdown precedes retinal degeneration in age-related macular degeneration.

This work presents a combined light and electron microscopical approach to investigate the initial breakdown of the retinal pigment epithelium (RPE) and choriocapillaris (CC) in age-related macular degeneration (AMD). Perimacular sections of 12 dry and wet AMD eyes (82 ± 15 years) and 7 age-matched controls (75 ± 10 years) without retinal pathology were investigated. Disease progression was classified into 5 stages of retinal degeneration to investigate the concurrent CC breakdown. Special emphasis was laid on transitions where intact CC-RPE-retina complexes went over into highly atrophied areas. AMD sections showed elevated loss of photoreceptors, RPE and CC (p < 0.01), and thickened Bruch's membrane with increased basal laminar and linear deposits compared with controls. Up to 27% of the CC was lost in controls although RPE and retina were still intact. This primary loss of CC further increased with AMD (up to 100%). The data implicate that CC breakdown already occurs during normal aging and precedes degeneration of the RPE and retina with AMD, defining AMD as a vascular disease. Particular attention should be given to the investigation of early AMD stages and transitional stages to the late stage that reveal a possible sequence of degenerative steps with aging and AMD.

Towards a quantitative OCT image analysis.

BACKGROUND: Optical coherence tomography (OCT) is an invaluable diagnostic tool for the detection and follow-up of retinal pathology in patients and experimental disease models. However, as morphological structures and layering in health as well as their alterations in disease are complex, segmentation procedures have not yet reached a satisfactory level of performance. Therefore, raw images and qualitative data are commonly used in clinical and scientific reports. Here, we assess the value of OCT reflectivity profiles as a basis for a quantitative characterization of the retinal status in a cross-species comparative study. METHODS: Spectral-Domain Optical Coherence Tomography (OCT), confocal Scanning-Laser Ophthalmoscopy (SLO), and Fluorescein Angiography (FA) were performed in mice (Mus musculus), gerbils (Gerbillus perpadillus), and cynomolgus monkeys (Macaca fascicularis) using the Heidelberg Engineering Spectralis system, and additional SLOs and FAs were obtained with the HRA I (same manufacturer). Reflectivity profiles were extracted from 8-bit greyscale OCT images using the ImageJ software package (http://rsb.info.nih.gov/ij/). RESULTS: Reflectivity profiles obtained from OCT scans of all three animal species correlated well with ex vivo histomorphometric data. Each of the retinal layers showed a typical pattern that varied in relative size and degree of reflectivity across species. In general, plexiform layers showed a higher level of reflectivity than nuclear layers. A comparison of reflectivity profiles from specialized retinal regions (e.g. visual streak in gerbils, fovea in non-human primates) with respective regions of human retina revealed multiple similarities. In a model of Retinitis Pigmentosa (RP), the value of reflectivity profiles for the follow-up of therapeutic interventions was demonstrated. CONCLUSIONS: OCT reflectivity profiles provide a detailed, quantitative description of retinal layers and structures including specialized retinal regions. Our results highlight the potential of this approach in the long-term follow-up of therapeutic strategies.

Effects of a single intravitreal injection of aflibercept and ranibizumab on glomeruli of monkeys.

PURPOSE: It is known that endothelial cells in the kidney are also strongly VEGF-dependent. Whether intravitreal drugs can be detected within the glomeruli or affect VEGF in glomerular podocytes is not known. Therefore, the aim of this pilot study was to investigate the effects of a single intravitreal injection of aflibercept and ranibizumab on glomeruli of monkeys. METHODS: The kidneys of eight cynomolgus monkeys, which were intravitreally injected either with 2 mg of aflibercept or with 0.5 mg of ranibizumab, were investigated one and seven days after injection. Two animals served as controls. The distribution of aflibercept, ranibizumab and VEGF was evaluated using anti-Fc- or anti-F(ab)-fragment and anti-VEGF antibodies respectively. The ratio of stained area/nuclei was calculated using a semi-quantitative computer assisted method. Glomerular endothelial cell fenestration was quantified in electron microscopy using a systematic uniform random sampling protocol and estimating the ratio of fenestrae per µm. RESULTS: Compared to the controls, the anti-VEGF stained area/nuclei ratio of the ranibizumab-treated animals showed no significant changes whereas the stained areas of the aflibercept-treated monkeys showed a significant decrease post-treatment. Immune reactivity (IR) against aflibercept or ranibizumab was detected in aflibercept- or ranibizumab treated animals respectively. The number of fenestrations of the glomerular endothelial cells has shown no significant differences except one day after aflibercept injection in which the number was increased. CONCLUSION: Surprisingly, both drugs could be detected within the capillaries of the glomeruli. After a single intravitreal injection of aflibercept, VEGF IR in the podocytes was significantly reduced compared to controls. Ranibizumab injection had no significant effect on the glomeruli's VEGF level. Whether this is caused by aflibercept's higher affinity to VEGF or because it is used in a higher stoichiometric concentration compared to ranibizumab remains to be investigated.

Effects of bevacizumab in retina and choroid after intravitreal injection into monkey eyes.

OBJECTIVE: Due to its low price, bevacizumab, which binds vascular endothelial growth factor, is currently used off-label for the treatment of over 50 different eye diseases and has been adopted worldwide despite the absence of serious preclinical data. This study examines the effects of intravitreal bevacizumab on monkey eyes with particular focus on choroidal and retinal vessels. METHODS: Cynomolgus monkeys received an intravitreal injection of 1.25 mg bevacizumab with or without (125)I labeling. The eyes were enucleated between 1 and 14 days after injection and were investigated by electron microscopy, immunocytochemistry, histochemistry or autoradiography. Untreated and phosphate buffered saline (PBS)-injected monkeys were used as controls. RESULTS: Bevacizumab locally accumulated at high concentration within individual blood vessels. It formed electron-dense deposits inside retinal veins and between red and white blood cells, activated thrombocytes and induced retinal vein thrombosis. Retinal cells like Müller cells, astrocytes and microglia were also activated. High amounts of bevacizumab were found in retinal and choroidal vessels which may interfere with blood flow. CONCLUSIONS: The deposits on the retinal vein walls may provide a mechanistic basis for the observed retinal blood flow alterations after bevacizumab treatment in patients.

In vitro induction of protein complexes between bevacizumab, VEGF-A¹65 and heparin: explanation for deposits observed on endothelial veins in monkey eyes.

PURPOSE: By investigating the effects of intravitreal bevacizumab on retinal vessels of monkeys, we found that bevacizumab accumulated locally at high concentration within individual blood vessels. It formed electron-dense fibrous deposits between endothelial cells and erythrocytes or granulocytes inducing retinal vein thrombosis. To better characterise the observed deposits, we investigated in vitro whether these deposits result from a complex between bevacizumab, vascular endothelial growth factor (VEGF)-A(165) and heparin. METHODS: Cynomolgus monkeys were intravitreally injected with 1.25 mg bevacizumab. The eyes were enucleated between 1 and 14 days after injection and investigated by electron microscopy and immunohistochemistry. Human umbilical vein endothelial cells (HUVEC) were incubated with bevacizumab, VEGF-A(165) and heparin at different concentrations. Treatments with ranibizumab served as control. Bevacizumab and ranibizumab were detected immunohistochemically using Cy-3 or immunogold labelled antibodies. RESULTS: Treated animals showed bevacizumab locally at high concentration within retinal blood vessels. Electron-dense deposits inside retinal vessels and between erythrocytes were detected in three out of four treated monkeys. In vitro, many globular aggregates heavily stained with anti-human IgG were only observed with equimolar amounts (240 nM) of bevacizumab and VEGF-A(165) and 0.2 U/ml heparin and not after ranibizumab treatment. The immunogold labelling specifically localised ultrastructurally the complexes formed between bevacizumab, VEGF-A(165) and heparin at the surfaces of HUVEC cells. CONCLUSIONS: Heparin promotes bevacizumab immune complex deposition on to endothelial cells. Our in vitro results could explain the presence of deposits observed on endothelial veins in monkey eyes intravitreally injected with bevacizumab.