Search

Aberrant growth of blood vessels in the choroid layer of the eye, termed choroidal neovascularization (CNV), is the pathological hallmark of exudative age-related macular degeneration (AMD), causing irreversible blindness among the elderly. Co-localization of proangiogenic factors and hypoxia inducible factors (HIF) in neovascular membranes from AMD eyes suggests the role of hypoxia in pathogenesis of CNV. In order to utilize hypoxic conditions in RPE for therapeutic purposes, we developed an optimized hypoxia regulated, RPE cell-specific gene therapy to inhibit choroidal neovascularization. An adeno-associated virus (AAV2) vector comprising a RPE-specific promoter and HIF-1 response elements (HRE) was designed to regulate production of human endostatin (a powerful angiostatic protein) in RPE. The vector was tested in a mouse model of laser-induced CNV using subretinal delivery. Spectral domain optical coherence tomography (SD-OCT) images from live mice and confocal images from lectin stained RPE flat mount sections demonstrated reduction in CNV areas by 80% compared to untreated eyes. Quantitative real-time polymerase chain reaction (qPCR) confirmed exogenous endostatin mRNA expression from the regulated vector that was significantly elevated 3, 7, and 14 days following laser treatment, but its expression was completely shut off after 45 days. Thus, RPE-specific, hypoxia-regulated delivery of anti-angiogenic proteins could be a valuable therapeutic approach to treat neovascular AMD at the time and in the ocular space where it arises.An optimized gene therapy vector targeting hypoxia and tissue-specific expression has been designed. The inhibitory role of gene therapy vector was tested in a mouse model of laser-induced CNV. An 80% reduction in choroidal neovascularization was achieved by the optimized vector. The expression of endostatin was limited to retinal pigment epithelium and regulated by hypoxia.

The retinal pigmented epithelial (RPE) layer is one of the major ocular tissues affected by oxidative stress and is known to play an important role in the etiology of age-related macular degeneration (AMD), the major cause of blinding in the elderly. In the present study, sulindac, a nonsteroidal antiinflammatory drug (NSAID), was tested for protection against oxidative stress-induced damage in an established RPE cell line (ARPE-19). Besides its established antiinflammatory activity, sulindac has previously been shown to protect cardiac tissue against ischemia/reperfusion damage, although the exact mechanism was not elucidated. As shown here, sulindac can also protect RPE cells from chemical oxidative damage or UV light by initiating a protective mechanism similar to what is observed in ischemic preconditioning (IPC) response. The mechanism of protection appears to be triggered by reactive oxygen species (ROS) and involves known IPC signaling components such as PKG and PKC epsilon in addition to the mitochondrial ATP-sensitive K(+) channel. Sulindac induced iNOS and Hsp70, late-phase IPC markers in the RPE cells. A unique feature of the sulindac protective response is that it involves activation of the peroxisome proliferator-activated receptor alpha (PPAR-α). We have also used low-passage human fetal RPE and polarized primary fetal RPE cells to validate the basic observation that sulindac can protect retinal cells against oxidative stress. These findings indicate a mechanism for preventing oxidative stress in RPE cells and suggest that sulindac could be used therapeutically for slowing the progression of AMD.

PURPOSE Muller cells, the major glial cell in the retina, play a significant role in retinal neovascularization in response to tissue hypoxia. We previously designed and tested a vector using a hypoxia-responsive domain and a glial fibrillary acidic protein (GFAP) promoter to drive green fluorescent protein (GFP) expression in Muller cells in the murine model of oxygen-induced retinopathy (OIR). This study compares the efficacy of regulated and unregulated Muller cell delivery of endostatin in preventing neovascularization in the OIR model. METHODS Endostatin cDNA was cloned into plasmids with hypoxia-regulated GFAP or unregulated GFAP promoters, and packaged into self-complementary adeno-associated virus serotype 2 vectors (scAAV2). Before placement in hyperoxia on postnatal day (P)7, mice were given intravitreal injections of regulated or unregulated scAAV2, capsid, or PBS. Five days after return to room air, on P17, neovascular and avascular areas, as well as expression of the transgene and vascular endothelial growth factor (VEGF), were compared in OIR animals treated with a vector, capsid, or PBS. RESULTS The hypoxia-regulated, glial-specific, vector-expressing endostatin reduced neovascularization by 93% and reduced the central vaso-obliteration area by 90%, matching the results with the unregulated GFAP-Endo vector. Retinas treated with the regulated endostatin vector expressed substantial amounts of endostatin protein, and significantly reduced VEGF protein. Endostatin production from the regulated vector was undetectable in retinas with undamaged vasculature. CONCLUSIONS These findings suggest that the hypoxia-regulated, glial cell-specific vector expressing endostatin may be useful for treatment of neovascularization in proliferative diabetic retinopathy.

Wolfram syndrome 1 (WFS1, OMIM 222300), a rare genetic disorder characterized by optic nerve atrophy, deafness, diabetes insipidus and diabetes mellitus, is caused by mutations of WFS1, encoding WFS1/wolframin. Non-syndromic WFS1 variants are associated with the risk of diabetes mellitus due to altered function of wolframin in pancreatic islet cells, expanding the importance of wolframin. This study extends a previous report for the monkey retina, using immunohistochemistry to localize wolframin on cryostat and paraffin sections of human retina. In addition, the human retinal pigment epithelial (RPE) cell line termed ARPE-19 and retinas from both pigmented and albino mice were studied to assess wolframin localization. In the human retina, wolframin was expressed in retinal ganglion cells, optic axons and the proximal optic nerve. Wolframin expression in the human retinal pigment epithelium (RPE) was confirmed with intense cytoplasmic labeling in ARPE-19 cells. Strong labeling of the RPE was also found in the albino mouse retina. Cryostat sections of the mouse retina showed a more extended pattern of wolframin labeling, including the inner nuclear layer (INL) and photoreceptor inner segments, confirming the recent report of Kawano et al. [Kawano, J., Tanizawa, Y., Shinoda, K., 2008. Wolfram syndrome 1 (Wfs1) gene expression in the normal mouse visual system. J. Comp. Neurol. 510, 1-23]. Absence of these cells in the human specimens despite the use of human-specific antibodies to wolframin may be related to delayed fixation. Loss of wolframin function in RGCs and the unmyelinated portion of retinal axons could explain optic nerve atrophy in Wolfram Syndrome 1.

Mouse retinal photoreceptor cell generation and morphogenesis take place in a well-characterized temporal sequence. Both rod and cone photoreceptor differentiation and synaptogenesis occur postnatally, but the relative timing of these events has been difficult to document due to the paucity of cell-specific markers. We have found that antibodies to neuron-specific enolase (NSE) preferentially label a subpopulation of photoreceptors in the outer nuclear layer (ONL) of the mouse retina in addition to labeling ganglion, amacrine, bipolar, and horizontal cells within the inner layers of the retina. The appearance of NSE immunoreactivity in the different classes of retinal neurons during development showed a close temporal relationship to the onset of expression of the synaptic vesicle-associated protein SV2 and clearly preceded the sequential development of synaptic connections in both inner and outer synaptic layers. The NSE-immunoreactive photoreceptors were identified as cones by dual labeling of their inner segments with the lectin peanut agglutinin or by colabeling with antisera to cone photopigments. Axonal extensions of NSE-labeled cone cells were shown to interact with those of differentiating horizontal cells as early as postnatal day 3 (P3). Colocalization of NSE with SV2 indicated that cone cells began to make synaptic contacts with horizontal cell processes several days prior to the development of rod synaptic terminals. Between P4 and P11, cone photoreceptor cell nuclei were observed to be scattered at various levels throughout the ONL and thus appeared to have become displaced from their previous position directly beneath the outer limiting membrane (OLM). By P12, the cone nuclei had migrated sclerad once again and were now observed to be neatly aligned adjacent to the OLM. In the rd mouse mutant, this migratory process was delayed, so that, at P12, positioning of the cone cell nuclei within the ONL was still quite irregular. Thus, we have identified a late migratory phase for cone photoreceptors during the second week after birth that correlates with the timing of maturation of the rod synaptic terminals just prior to eye opening. The types of cues used by maturing cone cells for their eventual sclerad location remain to be elucidated. J. Comp. Neurol. 388:47–63, 1997. © 1997 Wiley-Liss, Inc.

Selective degeneration of rod photoreceptor cells in the retinal degenerative (rd) mouse prior to their complete maturation is thought to result from elevated cyclic guanosine monophosphate (cGMP) levels owing to the inherited defect in cGMP-phosphodiesterase. To investigate potential signaling pathways which might lead to apoptotic death of photoreceptors in the rd retina, the expression of immediate-early genes (IEG) of the activating protein-1 transcription factor (AP-1) family was examined. Increasing numbers of apoptotic photoreceptor nuclei were observed in the outer nuclear layer of the rd mouse beginning at postnatal day (P) 10. The peak incidence of apoptotic cells was observed at P13; by P16, almost the entire population of photoreceptors had been lost. Although c-Fos-like immunoreactivity was absent in photoreceptors of normal retinas, we observed that commencing at around P10, increasing numbers of rod photoreceptors in the rd retina exhibited nuclear staining for c-Fos protein. While no change in the distribution patterns of other members of the AP-1 family (c-Jun, JunB, and JunD) was observed in photoreceptors, Muller cell nuclei were transiently immunoreactive for c-Jun on P11. The incidence of c-Fos-positive photoreceptors peaked sharply at P12, 1 day earlier than the peak in apoptosis. Furthermore, the population of c-Fos-positive photoreceptors was distinct from apoptotic photoreceptors exhibiting chromatin condensation. The aberrant expression of c-Fos protein in rod photoreceptors immediately prior to their death in the rd mouse raises the possibility that c-Fos may be directly or indirectly involved in triggering the apoptotic cascade. Furthermore, the additional finding of c-Jun induction in Muller glia suggests that the IEG response to photoreceptor degeneration involves both intra- and intercellular signal transduction pathways.

Purpose. A retroviral marker was used to label daughter cells arising from individual neuroblasts in the rd mouse retina, in order to investigate the hypothesis that a clonal relationship exists among degenerating photoreceptor cells. METHODS. On the day of birth, a single injection of retrovirus with a lac Z (s-galactosidase) reporter construct was injected into the retina in the vicinity of the subretinal space. Descendants of single neuroblasts were identified histochemically by examining the retinas at P14 (postnatal day 14). Light and electron microscopic studies were used to identify the retrovirally-induced marker s-D-galactosidase using Bluo-gal dye. Double-labeling of degenerating cone cells was accomplished by taking 100 mm vibratome sections of retrovirally-injected eyes and using either FITC-PNA or HRP-PNA to visualize clusters of degenerating cones as well as Bluo-gal labeled clones of photoreceptor cells on the same tissue section. RESULTS. A relatively large number of clones of primarily ph...

Lens epithelial cells undergo a dramatic transformation during the process of differentiation into elongated fiber cells. The membrane-associated actin cytoskeleton is likely to play a critical role in the stabilization and maintenance of the highly elongated fiber cell shape. Tropomodulin is a tropomyosin-binding protein associated with actin filaments in a variety of terminally differentiated cell types where stable actin filament organization is required for cell function. We now present results of studies to determine the temporal expression of tropomodulin in the developing lens. In situ hybridization experiments detected expression of tropomodulin mRNA in the developing mouse lens in elongating cells with a pattern similar to that of the fiber specific beta- and gamma-crystallins. Tropomodulin mRNA expression first appeared around 11.5 days post-coitum in elongating cells in the posterior part of the lens vesicle. At later stages the signal for tropomodulin was present in the elongating cells at the lens equator and in cortical fiber cells; signal was absent from the epithelium. To investigate the possible link between tropomodulin expression and fiber differentiation we used a well-established lens epithelial explant culture system in which fiber differentiation is induced by fibroblast growth factor (FGF). Tropomodulin expression was only observed in FGF-treated explants in conjunction with morphologic changes characteristic of lens fiber cell differentiation. The appearance of tropomodulin during the process of fiber cell differentiation suggests that tropomodulin may be important for stabilization and/or determination of actin filament length.

Morphometric analysis of the numbers of neurons in the ganglion cell layer (GCL) of the central retina (fovea/foveola/parafoveal retina) in eyes from 9 Alzheimer's disease (AD) and 11 age-matched control cases revealed an overall decrease of 25% in total numbers of neurons in AD as compared with control eyes. Detailed analyses of GCL neurons at various eccentricities from the foveola showed that the greatest decrease in neuronal density (43% decrease) occurred in the central 0-0.5 mm (foveal region), while at 0.5-1 mm and at 1-1.5 mm eccentricities, neuronal loss amounted to 24 and 26%, respectively. The temporal region of the central retina appeared most severely affected, with up to 52% decrease in neuronal density near the foveola (central 0-0.5 mm eccentricity). There was close agreement between fellow eyes analyzed separately for three AD and three control cases. Analysis of neuronal sizes showed that all sizes of neurons were similarly affected in AD. In the GCL of control retinas, neurons decreased with age (coefficient of correlation = -0.67), while in AD retinas no such relationship was evident. Since in the central 0-2 mm region of the retina 97% of neurons in the GCL are ganglion cells (while the remaining 3% consist of displaced amacrine cells), these results demonstrate extensive ganglion cell loss in the central retina in AD.

Somatostatin (SRIF) is a neuroactive peptide that is distributed throughout the nervous system, including the retina. This peptide has been localized to populations of amacrine cells in a variety of vertebrate species. In the rabbit retina, SRIF immunoreactivity is present in a sparse population of medium to large neurons (13.72 microns in diameter, or 147.84 mu 2) in the ganglion cell layer and in a small number of neurons in the inner nuclear layer. These cells display a preferential distribution to the inferior retina, with the highest density near the ventral and ventrolateral retinal margins (11.33 cells/mm2). SRIF-immunoreactive cells have two to five primary processes that arborize in the proximal inner plexiform layer (IPL). These give rise to a plexus of finer processes in the distal IPL. Occasional immunoreactive processes are also present in the outer plexiform layer. In the IPL, these laminar networks are present in all retinal regions. In addition, SRIF-immunoreactive cells often have a fine-caliber axonlike process that eminates from the soma or perisomal region. These processes travel for great distances across the retina in either the nerve fiber layer or in the distal IPL but are never seen to enter the optic nerve head. In addition, the number of SRIF-immunoreactive somata remains unchanged following transection of the optic nerve. Taken together, these data indicate that SRIF-immunoreactive neurons of the rabbit retina are displaced amacrine cells. Furthermore, the sparse distribution of SRIF-immunoreactive somata, the wide-ranging, asymmetric arborization of their cellular processes, and previous pharmacological studies suggest that these neurons mediate a broad modulatory role in retinal function.

Gliomas are the most common brain tumor in the central nervous system (CNS). The majority of malignant gliomas arise from neoplastic transformation of resident astrocytes. Gliomas are very aggressive tumors since they are characterized by widespread invasion of brain tissue. The exact pathogenesis and underlying mechanisms for glioma cell infiltration are currently unclear. Cell-cell interaction and tissue microenvironment play an important role in tumor progression leading to modification and infiltration of surrounding tissue. The hypoxic microenvironment contributes to abnormal neovascularization of the glioma.

Age-related macular degeneration (AMD) is the leading cause of blindness in the western world for people over 60 years of age. The most severe pathological event of AMD is choroidal neovascularization (CNV), the process of new vessel formation emerging from the choroid. The new vessels extend into the normally avascular photoreceptor cell layer, where they leak fluid and cause photoreceptor cell death. CNV is thought to be initiated by hypoxia and chronic inflammation, which occur due to abnormal, age-related changes within the retinal pigmented epithelium (RPE). These events cause increased expression of the angiogenic protein vascular endothelial growth factor (VEGF) via hypoxia-inducible factor-1 (HIF-1), a transcription factor that is vital in regulation of cellular responses to hypoxic and inflammatory conditions. Increased VEGF signaling stimulates proliferation and migration of vascular endothelial cells and facilitates the neovascular process.

Peanut agglutinin, a lectin with high binding affinity for galactose-galactosamine disaccharide, was used to monitor changes in the photoreceptor cell layer of mice with inherited retinal degeneration. Mice homozygous for the retinal degeneration (rd) gene exhibit a rapid loss of rod photoreceptor cells in the first postnatal month. Previous studies have shown that aggregates of peanut agglutinin-binding cells are observed in the outer nuclear layer in the retinal degenerative mouse at between postnatal days 10 and 18, a period during which massive photoreceptor degeneration occurs in this mutant. This study was performed to determine whether these peanut agglutinin-positive cell clusters represent degenerating photoreceptor cells or, alternatively, macrophages that have migrated into the photoreceptor cell layer. Electron microscopic cytochemistry, using horseradish-peroxidase-conjugated peanut agglutinin, was used to trace cellular processes of peanut-agglutinin-stained cell clusters. Additionally, macrophage-specific antibodies were employed to determine whether macrophages were present in the clusters. The cell clusters did not react with macrophage-specific antibodies. However, processes of cells in peanut-agglutinin-bound clusters could be traced by electron microscopic serial sections to both the outer limiting membrane and the outer synaptic layer. These results provide strong evidence that peanut-agglutinin-bound cells seen during this stage of degeneration in the rd mouse are degenerating photoreceptor cells. Since peanut agglutinin has been shown to bind preferentially to cone, but not to rod, photoreceptor cells, the results also suggest that the clusters may be aggregates of degenerating cones.

The Purkinje cell degeneration (pcd) mutant mouse rapidly loses cerebellar Purkinje cells and about 50% of its retinal photoreceptor cells at between 3 and 5 weeks of age, and thereafter slowly loses the remaining photoreceptor cells during the first year of life. An ultrastructural study of the developing photoreceptor cells of the pcd/pcd retina was undertaken using both transmission and scanning electron microscopy to characterize further the previously reported retinal vesicles associated with this mutation. Transmission electron microscopy (TEM) revealed an abundance of 'bead-like' vesicles or excrescences in the extracellular matrix surrounding the inner segment region at post-natal day (P) 25. The vesicles are membrane bound, amorphous in appearance and vary in size from 125 to 370 nm. Scanning electron microscopy suggests that the vesicles seen with TEM are actually spherules formed from outpocketing and pinching off of the plasma membrane in the mitochondria-rich region of the rod inner segment. At P25, the spherules are concentrated in the interphotoreceptor space at the level of rod inner segments; at P40, however, they are displaced from their origin and appear mostly at the level of rod outer segments and in the subretinal space.

PURPOSE. Metallothioneins (MTs) in the brain and retina are believed to bind metals and reduce free radicals, thereby protecting neurons from oxidative damage. This study was undertaken to investigate whether retinal photoreceptor (PR) cells lacking MTs are more susceptible to hyperbaric oxygen (HBO)–induced cell death in vivo. METHODS. Wild-type (WT) and MT-knockout (MT-KO) mice lacking metallothionein (MT)-1 and MT-2 were exposed to three atmospheres of 100% oxygen for 3 hours, 3 times per week for 1, 3, or 5 weeks. The control animals were not exposed. Histologic analysis of PR viability was performed by counting rows of nuclei in the outer nuclear layer (ONL). Ultrastructure studies verified PR damage. RESULTS. HBO exposure produced a major loss of PR cells in the central retinas of WT and MT-KO mice, with no effect on the peripheral retina even at the longest (5 weeks) exposures. The degree of PR damage and cell death increased with duration of HBO exposure. One week of HBO exposure was insufficient to cause PR death, but tissue damage was observed in the inner and outer segments. At 3 weeks, the rows of PR nuclei in the central retina were significantly reduced by 38% in WT and 28% in MT-KO animals. At 5 weeks, PR loss was identical in WT (34%) and MT-KO (34%) animals and was comparable to that in WT at 3 weeks. CONCLUSIONS. The data suggest that MT-1 and -2 alone are not sufficient for protecting PRs against HBO-induced cell death. The selective degeneration of central PRs may provide clues to mechanisms of oxidative damage in retinal disease. (Invest Ophthalmol Vis Sci. 2008;49:3207–3215) DOI:10.1167/iovs.071039

Loss of pericytes is a critical event early in the progression of microvascular dysfunction in diabetic retinopathy. Pericyte loss may be linked to high glucose mediated reactive oxygen species generation, blocking N-cadherin trafficking to the endothelial cell surface preventing pericyte recruitment and vessel stabilization. Hydrogen peroxide has been identified as a major free radical produced during high glucose exposure in endothelial cells. The goal of this research is to determine if tissue-specific hypoxia-regulated expression of a mitochondria-targeted catalase can prevent or limit RF/6A microvascular endothelial cell apoptosis and decrease vascular permeability by limiting cellular oxidative stress. For the development of tissue-specific and hypoxia-enhanced expression vectors, promoters were constructed with nine tandem combinations of HREs. This 9x HRE oligomer enhancer was inserted together into a pGL3 firefly luciferase plasmid with the Tie2( short ) promoter for endothelial-specific expression. The 9xHRE-Tie2( sh ) promoter construct was highly selective for RF/6A cells producing a basal amount of mitochondria-targeted catalase equivalent to the Tie2( short ) promoter alone. In response to hypoxia ( pO 2 = 1% ), the 9xHRE-Tie2( short ) promoter showed a 21-fold hypoxia-inducible activation similar in strength to the CMV promoter , measured by dual luciferase assay. The hybrid promoters were incorporated into a replication deficient AAV delivery system for apoptosis and cell culture based endothelial permeability assays. In preliminary assays using RF/6A microvascular endothelial cells, apoptosis was reduced by 58% and permeability was reduced by 46%. The results suggest that mitochondria-targeted catalase protects RF/6A microvascular endothelial cells from apoptosis and reduces endothelial permeability in a high-glucose, low-oxygen environment.

To compare the transduction efficiency of a lentiviral vector in the retina of normal mice and retinal degenerative (rd) mice following subretinal injection at various postnatal ages.Subretinal injections of lentiviral vector (pHR-CMV-GFP, 107IU/ml) were performed in normal (C57/6J) and rd mice on postnatal days P1 to P7 using a trans-scleral method and on days P14-P35 by a trans-corneal method. One to six weeks later the eyes were prepared for histological analysis. GFP positive cells were identified in retinal sections and retinal whole mounts to determine the overall extent and distribution of lentiviral transduction.Expression of GFP was observed adjacent to the injection site starting about 1 week after injection in both normal and rd mice and lasted 6 weeks (the longest period examined). In normal mice, GFP expression continued to increase and peaked around 2-3 weeks after injection with expression varying from approximately one quarter to the entire retina. GFP expression peaked earlier in rd mice injected from P1 to P7 compared to normal mice. Lenti-GFP expression decreased rapidly in rd mice older than P15. This was attributed to a period of intensive photoreceptor (PR) degeneration characteristic to this mutant. Retinal GFP expression was virtually absent in eyes injected after P14 in both normal and rd mice. Histological sections from P3 injected eyes showed GFP expression 9 days post-injection in both retinal pigment epithelium (RPE) and photoreceptor (PR) cells. GFP expression in RPE cells was stronger than that in PR cells. Both rods and cones expressed the lenti-GFP. GFP expression was limited to the RPE of normal mice if injections were performed at P14 or later. In rd mice, GFP expression in RPE was observed one week after injection at P1; GFP+-PR and -RPE cells were first detected 9 days after injection at P1, and 7 days after injection at P3-P7; RPE cells and occasional Muller cells around the injection site were GFP+ when the injection was performed at P14 or later.Lentiviral-mediated GFP transduction of RPE was efficient and sustained at all ages examined in both the normal and rd mouse. Trans-scleral, subretinal injection of lenti-GFP during the first postnatal week produced age-dependent transduction of PR cells in both mouse strains. Lenti-GFP expression was absent in both mouse strains if injections occurred after P14. There was a dramatic decrease in the transduction efficiency in rd mouse retinas corresponding to the degeneration of PR cells. However, the early stages of retinal degeneration in rd mice appeared to increase the transduction efficiency of PR cells. These data suggest that both age and degree of PR degeneration are important parameters to consider when designing gene therapy experiments or protocols.

Detailed analyses of neuronal and astrocyte cell numbers in the ganglion cell layer (GCL) of whole-mounted peripheral retinas from 16 Alzheimer's disease (AD) and 11 control eyes (11 and 9 cases, respectively) demonstrate extensive neuronal loss throughout the entire retina in AD as compared to control eyes. The observed neuronal loss is most pronounced in the superior and inferior quadrants, ranging between 40 and 49% throughout the midperipheral regions, and reaching 50-59% in the far peripheral inferior retina, while the overall neuronal loss throughout the entire retina amounts to 36.4% (p0.004). Although the 16% increase in astrocyte numbers is not significant, the ratio of astrocytes to neurons is significantly higher (82%; p0.0008) in AD as compared to normal retina (0.238 +/- 0.070 vs. 0.131 +/- 0.042). These results are strengthened by the close agreement (within +/- 15% of respective means) found between fellow eyes. Analysis of glial fibrillary acidic protein immunoreactivity (GFAP-ir) in sections of retinas from an additional 12 AD and 19 control cases show increased GFAP-ir with more extensive labeling of astrocytes in the GCL as well as increased labeling of Müller cell end-feet and radial processes in AD as compared to control retinas. The extensive loss of neurons documented in these retinas, accompanied by an increased astrocyte/neuron ratio, provides further support for the substantial involvement of the retina in AD.

Muller cells have been proposed to play an important role in photoreceptor cell development during the final stages of retinal maturation. The effect of disrupting Muller cells during mouse retinal development was investigated using the specific glial cell toxin, DL-alpha-aminoadipic acid (AAA). By giving multiple systemic injections over several days, impairment of Muller cell function was maintained during the period of photoreceptor migration and differentiation. Following three consecutive days of AAA treatment [commencing on post-natal (P) day 3, 5, 7 or 9, and examined at P8-P14], clumps of photoreceptor nuclei were displaced through the inner segments, lying immediately beneath the retinal pigment epithelium (RPE). Apart from the scalloped appearance of the outer retina, the overall lamination pattern of the retina was relatively well preserved. Even when AAA treatment commenced as early as P3, several days prior to the formation of the outer nuclear layer, the majority of photoreceptors migrated to their correct position and formed inner and outer segments. Therefore, the signals for photoreceptor migration are either provided by the Muller cells prior to P3, or, alternatively, are derived from different intrinsic or extrinsic cues. Disruption of Muller cell function was evidenced by decreased glutamine synthetase activity as well as by increased glial fibrillary acidic protein (GFAP) and decreased cellular retinaldehyde-binding protein (CRALBP) immunoreactivity. Immunocytochemistry with an antibody to CD44, which labels the microvilli of Muller cells at the outer limiting membrane, coupled with electron microscopic analysis, demonstrated that the zonulae adherentes between Muller cells and photoreceptors were either irregular or absent in areas adjacent to displaced clumps of photoreceptors. Thus AAA treatment of early post-natal mice results in localized disruption of the contacts between Muller cells and photoreceptors. These pathologic changes persist into adulthood since at P28, while short stretches of photoreceptors appeared relatively normal with fully developed outer segments, periodic clumps of displaced photoreceptor nuclei were still present adjacent to the RPE. In conclusion, Muller cell processes at the outer limiting membrane appear to play a critical role in providing a barrier to aberrant photoreceptor migration into the subretinal space.

Gene therapy for the retina has been employed successfully in humans and animal models for the treatment of retinal dystrophies.1–3 The retina is attractive for gene therapy approaches because it is surgically approachable, isolated due to the presence of the blood-retina barrier and immunologically privileged. The requirements for successful gene therapy include efficient and sustained gene transfer and choice of a gene product that is capable of eliciting therapeutic efficacy.4 The potential value of cell-specific and regulated gene therapy for the eye has been proposed for models of AMD, photoreceptor degeneration, and retinal ischemia.5–8 Alterations in retinal oxygen availability can form a basis for disease-appropriate patterns of transgene expression either at early stages of oxygen deprivation due to tissue stress or damage and at later stages of disease associated with tissue ischemia and cell necrosis. Oxygen is critical for maintaining retinal function and reduction in oxygen levels serve as a trigger for pathologic effects underlying AMD and diabetic retinopathy.9 Hypoxia-induced changes in the retina can also serve as a trigger for activation of gene therapy vectors designed for regulating transgene expression in response to depleted oxygen levels.2,6 Such tight regulation of the expression from gene therapy vectors is likely to be particularly important in retinal tissue where there are numerous distinct cell types with differing abilities to tolerate stress from hypoxia or elevated reactive oxygen species. The sensing of cellular hypoxia depends on the action of a key oxygen-dependent sensing system involving the transcription factor hypoxia inducible factor (HIF)-1, a heterodimer formed between the constitutive and ubiquitously expressed monomers HIF-1-alpha and HIF-1-beta.10 In normoxia, transcription is prevented because HIF-1-alpha is modified by hydroxylation of a proline residue and then processed for ubiquitin-mediated proteasomal degradation.11 Under hypoxic conditions, however, HIF-1-alpha dimerizes with its partner HIF-1-beta and translocates to the nucleus for activation of gene transcription. Transcriptional activation by HIF-1 occurs through binding of the factor to hypoxia response elements (termed HREs) in regulatory domains of target genes.11,12 Therapeutic products synthesized by hypoxia-regulated vectors have included growth factors such as bFGF and VEGF, antioxidant components, antiangiogenic factors including angiostatin and proapoptotic components such as Bax.13–17 The promoters of such hypoxia-regulated therapeutic vectors are designed to include a regulatory domain which incorporates multiple hypoxia responsive elements (HREs) which are known to bind the transcription factor HIF-1. We and others have reported that multimers of the HRE drive enhanced levels of gene expression relative to a single HRE.6,18 For further control over basal levels of expression and of inducibility, we have previously incorporated a neuronal silencing element into the promoters to prevent “leaky” gene expression under normoxic conditions.6 HRE containing promoters which also contain silencer elements have been found to elicit inducibility in hypoxia of greater than 50-fold.6 The primary focus of this study is aimed at testing the hypothesis that a hypoxia-responsive domain can be employed for activating cell-specific expression in Muller cells under conditions of hypoxic stress in the pathologic retina. A promoter containing a hypoxia responsive domain together with a cell specific promoter would be expected to retain cell specificity while also being hypoxia-inducible. In responding to hypoxia, the promoter will have potential for activation during early phase hypoxia as well as later ischemic phases noted in the progression of such disease processes as AMD and diabetic retinopathy. For developing a retinal gene therapy approach, a variety of gene products may be especially appropriate for expression in this vector system including pro-survival kinases, antioxidant enzymes that enhance Muller cell viability, and secreted factors either for preventing angiogenesis or for eliciting neuroprotection. Our HRE-regulated, retinal glial cell–specific promoter could be incorporated to investigate new therapies to treat a range of animal models of eye diseases such as inherited photoreceptor degeneration, age related macular degeneration, diabetic retinopathy, and glaucoma. Glial fibrillary acidic protein (GFAP) is the major intermediate filament in astrocytes and other glial cells, including nonmyelinating Schwann cells and Muller cells (the major glial element in the retina).19–21 Expression of GFAP serves as a marker of developmental processes as well as an indicator of gliosis in response to injury.21–23 Elevated levels of GFAP expression in retinal glial cells have been observed in light-induced retinal damage and in a number of eye diseases including diabetic retinopathy and AMD. Identification of the upstream regulatory sequences of the GFAP gene has been valuable for achieving glial cell–specific transgene expression either in transgenic mice or when employing gene transfer vectors. The domain from bp −2163 to +47 of the human GFAP gene has been found to drive astrocyte-specific expression whereas an internally truncated −2163 GFAP promoter lacking bp −1488 to −132 loses its cell specificity. However a short gfaABC1D promoter consisting of enhancer domains plus a proximal promoter domain was found to retain astrocyte-specific expression in the brain, while demonstrating at least a twofold higher activity than −2163 GFAP.24 In transgenic mice, the astrocyte-specific GfaABC1D provides more regional specificity in the brain than other GFAP promoter domains.24 In this investigation, both in vitro and in vivo studies were used to confirm the applicability of these regulatory domains for gene therapy strategies. The complete promoter was incorporated into a recombinant self complementary AAV (scAAV) containing the HRE-GFAP promoter driving GFP. Primary cultures of Muller cells were transfected to demonstrate a lack of expression of the GFP transgene in normoxia and high level transgene expression in hypoxia. For in vivo studies, the murine model of oxygen-induced retinopathy (OIR) is widely used to study neovascularization.25 In this model, postnatal day 7 (P7) mouse pups are exposed to high oxygen until P12 then returned to room air for another 5 days. During the initial high oxygen phase, normal retinal blood vessels regress. During the regression phase, retinal cells produce several HIF-1–mediated proangiogenic factors. This leads to development of abnormal neovascularization in the retina by P17.26 The expression of HIF-1 in P17 hypoxic retina is 31-fold greater than in P17 normoxic retina.27 The OIR model is used to study hypoxia-regulated reporter gene expression in the mouse retina.5 As our promoter requires HIF-1 for its induction, we chose the OIR model to evaluate the activity of our regulated promoter for in vivo studies.

The adverse effects of monosodium glutamate (MSG) in embryonic chick retinas in culture were examined, as a function of embryonic age, by light and electron microscopy. From embryonic day 7 (E7) through E14, the presence of a low threshold concentration of glutamate in the incubating medium causes damage in the inner layers of the retina. Retinas from older embryos appear to be more susceptible to the destructive effects of MSG. As reported earlier for E12 retinas ( Reif-Lehrer, Bergenthal and Hanninen, 1975 ), the lesion appears to be in the vitreal half of the inner nuclear layer with primarily amacrine and Muller cells being affected; with longer incubation periods ganglion cells also sustain damage. While damage from MSG has been documented in retina and other areas of the CNS in rodents treated with MSG in vivo, this is the first in-depth electron microscopic study of MSG damage to the isolated embryonic chick retina.

Observations are presented of a combined light and electron microscopic study of synaptogenesis in the photoreceptor terminal of visuallycompetent mice. Our study suggests that photoreceptor synaptogenesis in these retina progresses through a sequence of postnatal changes which leads to the ordered arrangement of processes within the adult photoreceptor terminal. On the 5th postnatal day, neuronal processes from synaptic contacts with a photoreceptor terminal, usually in the vicinity of a synaptic ribbon. Following the formation of synaptic contacts, the processes invaginate into the photoreceptor terminal, maintaining their position, lateral to a synaptic ribbon. Beginning on about the 10th day, a central process enters the terminal to yield a triad configuration which is a characteristic of a mature photoreceptor terminal. The formation of triads is essentially complete by the 14th postnatal day. Our observations suggest that horizontal cell processes form synaptic contacts with the photoreceptor terminal before bipolar cell dendrites, implying that the horizontal cells mature in advance of the bipolar cells. This possibility is discussed in terms of the appearance of the electroretinogram (ERG) in the developing mouse retina.

The effect of intraocular liquid silicone on the electroretinogram (ERG) and on retinal morphology was studied in rabbits. After vitrectomy, liquid silicone (1000 centistokes) or balanced salt solution (BSS) was injected into the eyes. The intraocular silicone was well tolerated clinically in all eyes that were followed over a period of 3 days to 6 months. ERG responses were equivalent in operated control and silicone-injected eyes. Light and electron microscopy showed slight but comparable changes in both operated control and silicone-injected eyes that were consistent with either fixation artifact and/or surgical trauma. The results suggest that exposure to silicone for up to 6 months does not have a toxic effect on the rabbit retina.

The Purkinje cell degeneration (pcd) mutant mouse rapidly loses cerebellar Purkinje cells between 3 and 5 weeks after birth and slowly loses retinal photoreceptor cells during the first year of life. In the present study, the retinal degeneration in the pcd mouse was analyzed by light microscopy and autoradiography throughout the first 15 months of age. By day 25 there is an abundance of pyknotic photoreceptor nuclei and many outer segments are clearly disorganized. Thereafter, as the photoreceptor cells are lost, their outer segments slowly become shorter and more variable in length; this slow change in length is explained by an almost proportional reduction in both rod outer segment renewal and disc shedding rates. At about 2 months of age, the rate of rod outer segment renewal is slightly less than half that in littermate controls, and the number of large phagosomes in pigment epithelial cells during the burst of disc shedding soon after light onset is one-half or less than that seen in littermate controls. Between 2 and 10.5 months of age, the retina in the inferior hemisphere of the eye shows substantially more-advanced photoreceptor degeneration than does the superior hemisphere, particularly in the far peripheral retina. A central-to-peripheral gradient of degeneration is conspicuous in the superior hemisphere; a similar but less obvious gradient of degeneration is also seen in the inferior hemisphere of the eye. Loss of photoreceptor cells and their synaptic terminals results in a predictable thinning of the outer synaptic layer. However, the inner nuclear layer shows no measurable thinning and the inner synaptic layer is reduced in thickness by only about 5-15%. Beginning at about 10 months of age, foci of thinned pigment epithelial cells are evident, and by 12 months there is some vascularization of the pigment epithelium by retinal capillaries.

Condensation and contraction of the vitreous have been implicated in the development of vitreoretinal traction and traction retinal detachment after penetrating ocular injury. In this study, using the inorganic dye ruthenium-red, we studied ultrastructural changes in vitreous in the rabbit eye after standard penetrating injury and intravitreal blood injection. In normal rabbit vitreous the collagen fibrils have a regular arrangement. In contrast, after a penetrating injury the collagen fibrils appear focally condensed. While it appears unlikely that such a network could alone produce tractional or directional forces, these alterations along the collagen fibrils could provide a scaffold along which contractile cells migrate into the vitreous.

Monoclonal antibody probes were used to identify antigenic cross reactivities among neuronal subpopulations and to dissect the human nervous system at several levels of organization. Six monoclonal antibodies, prepared with immunogens from Drosophila melanogaster or human nervous tissue, were used to localize antigens immunocytochemically in normal adult human neocortex, hippocampus, cerebellum, spinal cord, and retina. Four of the six antibodies were neural specific in their reactivity and each stained a unique combination of neurons. The antibodies reacted with at least three subpopulations of cerebral cortical neurons, including discrete populations of pyramidal and nonpyramidal cells. Components of a widely distributed functional system within the spinal cord and cerebellum were labelled by one antibody, which reacted with neurons in the nucleus dorsalis of Clarke, deep cerebellar nuclei, and Purkinje cells. At the single-cell level, three of the monoclonals differentially labelled the photoreceptor cell outer segment, inner segment, and perikaryon. Three of the six antibodies were reactive with specific protein bands, on immunoblots of tissue homogenates. This monoclonal antibody panel provides a novel and potentially useful method of analysis of the organization of the normal and diseased human nervous system.

Alzheimer's disease is a dementing disorder of unknown cause in which there is degeneration of neuronal subpopulations in the central nervous system. In postmortem studies, we found widespread axonal degeneration in the optic nerves of 8 of 10 patients with Alzheimer's disease. The retinas of four of the patients were also examined histologically, and three had a reduction in the number of ganglion cells and in the thickness of the nerve-fiber layer. There was no retinal neurofibrillary degeneration or amyloid angiopathy, which are typically seen in the brains of patients with Alzheimer's disease. The changes we observed in the patients with Alzheimer's disease were clearly distinguishable from the findings in 10 age-matched controls and represent a sensory-system degeneration that occurs in Alzheimer's disease. Study of the retina in patients with this disease may be helpful diagnostically, and isolation of the affected ganglion cells may facilitate molecular analysis of the disorder.

A battery of eight lectins with different carbohydrate specificities was used to study changes in glycoconjugate expression during cell differentiation in the mouse retina. The lectins tested included concanavalin A (Con A), wheat germ agglutinin (WGA), soybean agglutinin (SB A), peanut agglutinin (PNA), Ulex europaeus agglutinin (UEA), Ricinus communis agglutinin I (RCA), Dolichos biflorus agglutinin (DBA), and Limulus polyphemus agglutinin (LPA). Unfixed frozen sections of adult and early postnatal mouse retina were treated with fluorescein isothiocyanate-conjugated lectins and examined by fluorescence microscopy. The results showed selective lectin binding in both cellular and synaptic retinal layers of the adult mouse and throughout postnatal development. In general, an increase in intensity of fluorescent lectin staining during retinal development was observed for Con A, WGA, DBA, LPA, RCA, and PNA. This suggests an increase in the expression or accessibility of carbohydrate moieties during development. SBA and UEA showed little to no binding to adult or neonatal retina. Retinal vasculature was intensely stained by RCA, both during development and in the adult. All lectins binding to adult or neonatal retinal layers showed some degree of reactivity with the inner segment region of photoreceptor cells. However, only Con A, PNA and WGA bound to photoreceptor outer segments, suggesting significant differences in the glycosylated components of inner and outer segment membranes. PNA bound specifically to a subpopulation of photoreceptor cells and to discrete regions within the outer synaptic layer. The pattern of PNA binding suggests that this lectin binds preferentially to cone photoreceptor inner and outer segments and cone synaptic pedicles rather than to rod photoreceptor cells. This marked specificity of PNA binding suggests that it may provide a basis for the physical separation of cone and rod photoreceptor cells.

The use of acrylamide as an embedding medium for vertebrate retinal tissue and its applicability to lectin and antibody-based cytochemical studies is described. The acrylamide technique has numerous advantages over those using fresh-frozen or paraffin embedded material. The morphological integrity of retinal tissue prepared in acrylamide is equivalent to that obtainable with paraffin and superior to that of fresh-frozen material. In addition, this technique alleviates problems often encountered with the thermal and chemical treatments required in the paraffin method. The acrylamide technique allows the localization of lectin and antibody-binding sites essentially unaltered by the fixation and embedding protocol, as in frozen sections, while maintaining tissue morphology similar to that of paraffin-embedded tissue. It is hoped that this approach will be useful to other workers in vision research employing lectin, antibody or other cytochemical approaches to the study of cellular structure and function.

We have observed that the lectin Ricinus communis agglutinin I (RCA), which binds to terminal galactose moieties, serves as a marker for vasculature in the mouse retina. The binding of fluorescein-isothiocyanate-conjugated RCA was used to study the development of retinal vasculature in normal mice and in rd (retinal degeneration) mutant mice, which exhibit a massive loss of photoreceptor cells during the first month of life. In the normal mouse, retinal capillaries develop in an ordered manner and are concentrated in three major zones between the inner limiting membrane and the outer plexiform layer. In the rd mouse, the vessels appear to form normally but begin to degenerate by the end of the second postnatal week. By the end of the fourth postnatal week there is approximately a 35% reduction in the total number of vascular profiles in the rd retina compared to normal littermate controls. This reduction in vascularity is temporally associated with the photoreceptor degeneration.

Light microscopic autoradiographic localization of radioactive gamma-aminobutyric acid ([3H]GABA) accumulated by high-affinity uptake was compared in mouse retina after both in vitro and in vivo experiments. Retinas incubated 15 and 60 min with [3H]GABA showed selective glial accumulation of radioactivity with label over the entire Muller cell somata and processes. In contrast, intravitreal injection of [3H]GABA resulted in both neuronal and glial uptake, the characteristics of which appeared to be time-dependent. Label was predominantly confined to Muller cells 15 min after injection; when the post-injection period was lengthened to 60 min, label was found in both Muller and amacrine cells and in some somata in the ganglion cell layer. In addition to the diffuse labeling seen within the inner synaptic layer (ISL), a dense narrow band of silver grains appeared in the vitreal region of the ISL. The sites of accumulation of [3H]muscimol did not differ under in vivo and in vitro conditions.[3H]muscimol was always heavily concentrated in amacrine cell somata, in a few cell somata in the ganglion cell layer and seen diffusely in the ISL, with no correspondence to sites of amacrine-ganglion cell synapses. In vitro accumulation of [3H]GABA or [3H]muscimol was unaffected by the presence of GABA receptor-blocking drugs THIP (4,5,6,7-tetrahydroisoxazol-[5-4-c]pyridine-3-ol) and bicuculline and as blocked more effectively by diaminobutyric acid (DABA), an antagonist of neuronal uptake (Iversen and Kelly, 1975), than by beta-alanine, an antagonist of glial uptake (Schon and Kelly, 1975). Thus, both the accumulation of [3H]muscimol by amacrine cells and [3H]GABA by Muller cells showed the same pharmacological characteristics. These biochemical and autoradiographic results show that [3H]muscimol is not a good marker for postsynaptic GABA receptors when used under these in vivo and in vitro conditions.

In order to define cytoskeletal domains of the mammalian photoreceptor, actin and tubulin were localized in adult retinae of mouse and human. For light-microscopic localization, actin was labeled using fluorescent phalloidin or monoclonal antibodies against actin, and tubulin was labeled using monoclonal antibodies against alpha- and beta-tubulin in an immunocytochemical method. Actin and tubulin were also localized by ultrastructural immunocytochemistry in the mouse. Filamentous actin was present in the retina at the outer limiting membrane and in synaptic terminals, especially of the cones, while globular actin was observed additionally in the inner segments. Muller cell cytoplasm and apical microvilli at the outer limiting membrane were also labeled for filamentous actin. Alpha- and beta-tubulin were evident throughout the photoreceptors, including the inner segments, but not in the synaptic terminals or at the outer limiting membrane. In the early postnatal retina of mouse, actin and tubulin were present at the ventricular surface. This pattern changed as photoreceptors fully elongated and as synaptogenesis occurred in the outer plexiform layer.

This study documents the light-microscopic and ultrastructural characteristics of ganglion cell degeneration in the retinas of patients with Alzheimer's disease (AD). The results show degeneration in the retinal ganglion cells (RGCs) characterized by a vacuolated, 'frothy' appearance of the cytoplasm. The degeneration is unique in AD because of the absence of neurofibrillary tangles within the RGCs, or of neuritic plaques or amyloid angiopathy in the retinas or optic nerves of any of the cases examined. These results suggest that neuronal degeneration in the ganglion cell layer (GCL) should be added to the constellation of neuropathologic changes found in patients with Alzheimer's disease.

The cerebellar ataxia in Purkinje cell degeneration (pcd) mutant mice results from the rapid loss of Purkinje cells between 3 and 5 weeks after birth. The loss of photoreceptors in these mutants begins about the same time but proceeds slowly, with most photoreceptors being lost by 1 year of age. In this study the retinas of pcd/pcd mice and their littermate controls from the age of 10 postnatal days to 15 months were analyzed by electron microscopy. The first signs of photoreceptor cell degeneration are apparent in the region of photoreceptor inner segments as early as postnatal day 13, and more prominently at day 18. During this time, the degeneration is characterized by a large number of vesicles, ranging in diameter from 150 to 350 nm, which are located in the extracellular space adjacent to the photoreceptor inner segments. Analysis of serial sections shows that most of these membrane-bound degeneration profiles are tubular in shape and some are continuous with the cell membrane of the inner segment. Therefore, these "profiles" are thought to arise from tubular outpocketings of the inner segments which cleave off to form isolated membrane-bound profiles. This represents a new and unusual form of photoreceptor degeneration. While the most obvious abnormality in the retina is degeneration of photoreceptor cells, Muller cells also appear to be affected, with swollen apical processes often seen coursing through the outer nuclear layer.

The influence of photoreceptor degeneration upon synaptogenesis in the photoreceptor terminal of developing retinae of rd mice (C3H/HeJ and C57BL rd le) is outlined and compared to the sequential development of the photoreceptor terminals of the retinae of normal mice. The initial phases of synaptogenesis in the photoreceptor terminal of the rd retina appear normal during the first postnatal week. Appositional contacts form between the photoreceptor terminal and neuronal processes before invagination of horizontal processes occurs to form a dyad configuration. A virtual absence of triad configurations in the photoreceptor terminal suggests that synaptogenesis is arrested in the photoreceptor terminal of the rd retina between the 7th and 10th day, at which time the central (bipolar) process normally invaginates into the photoreceptor terminal. During this period, photoreceptor cell death proceeds rapidly but few signs of degeneration are apparent. It is suggested that the inherited abnormality influences the maturation of bipolar cells more than that of horizontal cells. The b-wave component of the electroretinogram (ERG) of the immature rd retina is discussed in terms of photoreceptor/bipolar cell synaptic contacts.

Incorporation of [3H]-cytidine into messenger RNAs (polyadenylated RNAs) was enhanced in light in isolated intact photoreceptor cells of the rat. The increase in polyadenylated (PolyA+)RNAs appeared selective relative to other photoreceptor RNAs since incorporation of [3H]-cytidine into this fraction was up to 10-fold higher while labeling of total cellular RNAs was only two-to three-fold higher in light compared with dark. The photoreceptor cells were isolated in vivo through destruction of inner retina neurons with injections of combinations of neurotoxic substances during early postnatal development. The photoreceptor cells attained normal adult morphology and function: the a-wave of the electroretinogram, as well as the thickness of the outer nuclear layer and the length of photoreceptor inner and outer segments, were found to be within the normal range at 4 and 10 weeks of age. In addition to RNA synthesis, such photoreceptor cell preparations when incubated in vitro demonstrated a capacity for regulating light-dependent sodium fluxes comparable to that within the intact retina. The potential usefulness of this model for exploring the molecular biology of photoreceptor cells is discussed.

Light microscopic autoradiography, following injection of tritiated thymidine, was used to study the postnatal development of cells in the inner nuclear layer (INL) of the normal and rd (C57BL/6J-rd le) mouse retina. Specific attempts were made to determine whether some bipolar cell degeneration occurs in conjunction with previously noted photoreceptor cell degeneration in the retina of the mutant (rd) mouse. A differential count of cell types in the INL of both the central and peripheral retina indicates that the same ratio of bipolar cells to Muller cells (3:1) exists in the normal and the rd retina during postnatal development. The density of cells in the INL was similar in both groups during development and similar decreases in width of the INL occurred in the normal and rd mouse during maturation. In addition, autoradiographic analysis of the proliferation of postnatally forming bipolar cells suggested that the rd gene is not lethal to the bipolar cell population which forms after birth. The labeling index (percent labeled cells) in the INL, determined as a function of the date of 3H-thymidine injection and assessed in the mature retina (day 41), was similar in both mice. Analysis of the number of silver grains/nucleus presented a more complete picture of how the progenitor pool of cells contributes to the cell types formed postnatally in the retinas of both mice. The conclusion that bipolar cell death does not occur to any significant degree in the retina of the rd also suggests that transneuronal degeneration does not occur in the retina of the mutant mouse during the period (up to day 41) when most of the photoreceptor cells die. The observation in both groups of mice that cell density remains constant between postnatal day 14 and day 41, while the width of the INL changes tremendously, suggests that there is a significant redistribution of cells during growth of the eyeball and/or cell death which occurs in all cell types found in the INL in both the normal and mutant mouse.

Posterior vitreous detachment (PVD) and epiretinal membranes occur in a number of vitreoretinal diseases. We have developed an experimental model in which we can provide the morphologic correlation of these dynamic processes. The method provides the opportunity to study epiretinal membrane formation with the scanning electron microscope (SEM); with SEM, some epiretinal membranes that could not be readily detected either clinically or by routine light microscopy can now be identified and studied in detail. We performed an experimental posterior penetrating injury with injection of autologous whole blood or blood and lens material into the vitreous. Five eyes with posterior vitreous detachment but no retinal detachment were selected for SEM. A reduction in the cortical vitreous filaments and the presence of epiretinal membranes was apparent with SEM. In most areas the epiretinal membranes were separated from the internal limiting membrane by a narrow cleft; however, limited attachment sites between the epiretinal membranes and retina were observed in areas overlying retinal blood vessels. In two eyes we observed microscopic retinal folds beneath the membranes, demonstrating a possible morphologic correlation between epiretinal cellular contraction and traction on the retina.

In the all-rod retina of the adult mouse, synaptic contacts in the outer plexiform layer occur between photoreceptor terminals and dendritic processes of horizontal and bipolar cells. The postnatal development of the outer plexiform layer has been studied by light and electron microscopy. Following decapitation and enucleation of the eyes, samples of retinae from DBA mice (newborn to fourteen days) were removed and fixed by immersion in solutions of buffered glutaraldehyde and paraformaldehyde. The tissue was postfixed in osmium tetroxide, dehydrated and embedded in araldite. Sections were stained with uranyl acetate and lead citrate.With light microscopy, the outer plexiform layer first appears in the central retina as a gap within the neuroblastic layer on the fifth postnatal day. The separation is completed by the seventh postnatal day and appears mature by the fourteenth postnatal day.