Intraorbital transection of the rabbit optic nerve: consequences for ganglion cells and neuroglia in the retina

Multifocal electroretinography increases following experimental glaucoma in nonhuman primates with retinal ganglion cell axotomy

PURPOSE

To determine whether short-latency changes in multifocal electroretinography (mfERG) observed in experimental glaucoma (EG) are secondary solely to retinal ganglion cell (RGC) loss or whether there is a separate contribution from elevated intraocular pressure (IOP).

METHODS

Prior to operative procedures, a series of baseline mfERGs were recorded from six rhesus macaques using a 241-element unstretched stimulus. Animals then underwent hemiretinal endodiathermy axotomy (HEA) by placing burns along the inferior 180° of the optic nerve margin in the right eye (OD). mfERG recordings were obtained in each animal at regular intervals following for 3-4 months to allow stabilization of the HEA effects. Laser trabecular meshwork destruction (LTD) to elevate IOP was then performed; first-order kernel (K1) waveform root-mean-square (RMS) amplitudes for the short-latency segment of the mfERG wave (9-35 ms) were computed for two 7-hexagon groupings-the first located within the superior (non-axotomized) macula and the second within the inferior (axotomized) macula. Immunohistochemistry for glial fibrillary acidic protein (GFAP) was done.

RESULTS

By 3 months post HEA, there was marked thinning of the inferior nerve fiber layer as measured by optical coherence tomography. Compared with baseline, no statistically significant changes in 9-35 ms K1 RMS amplitudes were evident in either the axotomized or non-axotomized portions of the macula. Following LTD, mean IOP in HEA eyes rose to 46 ± 9 compared with 20 ± 2 mmHg (SD) in the fellow control eyes. In the HEA + EG eyes, statistically significant increases in K1 RMS amplitude were present in both the axotomized inferior and non-axotomized superior portions of the OD retinas. No changes in K1 RMS amplitude were found in the fellow control eyes from baseline to HEA epoch, but there was a smaller increase from baseline to HEA + EG. Upregulation of GFAP in the Müller cells was evident in both non-axotomized and axotomized retina in eyes with elevated IOP.

CONCLUSIONS

The RMS amplitudes of the short-latency mfERG K1 waveforms are not altered following axotomy but undergo marked increases following elevated IOP. This suggests that the increase in mfERG amplitude was not solely a result of RGC loss and may reflect photoreceptor and bipolar cell dysfunction and/or changes in Müller cells.

Retrograde Maculopathy in Patients With Glaucoma

PURPOSE

Macular optical coherence tomography (OCT) analysis can be used for quantitative measures of optic nerve atrophy at a location far from the optic nerve head. This recently led to the finding of microcystic macular edema (MME), that is vacuolar inclusions in the macular inner nuclear layer, in some glaucoma patients. The involvement of individual retinal layers is yet unclear in glaucoma. In this study we systematically investigated glaucoma-induced changes in macular layers to evaluate whether glaucoma-associated damage extends beyond the macular ganglion cell layer.

PATIENTS AND METHODS

We included 218 consecutive patients and 282 eyes with confirmed primary open-angle glaucoma or pseudoexfoliation glaucoma, and macular OCT in a cross-sectional observational study. Eyes were screened for presence of MME. Thickness of individual retinal layers was determined using a semiautomatic segmentation algorithm. Peripapillary nerve fiber layer thickness and mean defect in visual field testing were extracted from OCT and medical records, respectively. Results were compared with a small group of eyes with no apparent glaucoma.

RESULTS

We found MME in 5 eyes from 5 primary open-angle glaucoma patients and 3 eyes of 3 pseudoexfoliation glaucoma patients (2.8%). MME was confined to the inner nuclear layer in a perifoveal ring and was associated with thinning of the ganglion cell layer and thickening of the macular inner nuclear layer. Glaucoma eyes without MME showed a significant inverse correlation of inner nuclear layer thickness with glaucoma severity.

CONCLUSIONS

Glaucomatous damage leads to a gradual thickening of the inner nuclear layer, which leads to MME in more severe glaucoma cases. These changes, along with nerve fiber loss and ganglion cell loss, may be summarized as glaucoma-associated retrograde maculopathy.

Increased Expression of Osteopontin in Retinal Degeneration Induced by Blue Light-Emitting Diode Exposure in Mice

Osteopontin (OPN) is a multifunctional adhesive glycoprotein that is implicated in a variety of pro-inflammatory as well as neuroprotective and repair-promoting effects in the brain. As a first step towards understanding the role of OPN in retinal degeneration (RD), we examined changes in OPN expression in a mouse model of RD induced by exposure to a blue light-emitting diode (LED). RD was induced in BALB/c mice by exposure to a blue LED (460 nm) for 2 h. Apoptotic cell death was evaluated by terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay. In order to investigate changes in OPN in RD, western blotting and immunohistochemistry were performed. Anti-OPN labeling was compared to that of anti-glial fibrillary acidic protein (GFAP), which is a commonly used marker for retinal injury or stress including inflammation. OPN expression in RD retinas markedly increased at 24 h after exposure, was sustained through 72 h, and subsided at 120 h. Increased OPN expression was observed co-localized with microglial cells in the outer nuclear layer (ONL), outer plexiform layer (OPL), and subretinal space. Expression was restricted to the central retina in which photoreceptor cell death occurred. Interestingly, OPN expression in the ONL/OPL was closely associated with microglia, whereas most of the OPN plaques observed in the subretinal space were not. Immunogold electron microscopy demonstrated that OPN was distributed throughout the cytoplasm of microglia and in nearby fragments of degenerating photoreceptors. In addition, we found that OPN was induced more acutely and with greater region specificity than GFAP. These results indicate that OPN may be a more useful marker for retinal injury or stress, and furthermore act as a microglial pro-inflammatory mediator and a phagocytosis-inducing opsonin in the subretinal space. Taken together, our data suggest that OPN plays an important role in the pathogenesis of RD.

Nestin is induced by hypoxia and is attenuated by hyperoxia in Müller glial cells in the adult rat retina

This study investigated the reactive changes in Müller glial cells and astrocytes of the rat retinae, which had been subjected either to hypoxia or to hypoxia followed by hyperoxia treatments. Fifteen rats were used. Ten rats were exposed to 9% O(2) for 2 h. Of these, five rats were killed at 24 h later; the remaining five rats were immediately exposed to 80% O(2) for 2 h and then killed 24 h later. Double immunofluorescence was carried out between nestin and glutamine synthetase (GS) and between glial fibrilary acidic proteins (GFAP) and GS in normal and pathological retinae. Enhanced nestin expression was observed in reactive astrocytes following hypoxia treatment as revealed in whole mount sections. A novel finding was the induction of nestin expression in Müller glial cells. Remarkably, the nestin immunostaining was downregulated to levels comparable to those of the normal rats with immediate hyperoxia treatment. Induced nestin expression by hypoxia colabelled with GFAP in astrocytes, however, remained unaffected after hyperoxia treatment. The induced expression of nestin in Müller glial cells and astrocytes in hypoxia and differential downregulation after hyperoxia treatment suggest a structural plasticity of the cytoskeletal framework of these cells. The differential response after hyperoxia treatment may be related to the functional states of the cells.

Involvement of gp130-associated cytokine signaling in Müller cell activation following optic nerve lesion

Ciliary neurotrophic factor (CNTF) and the related cytokine leukemia inhibitory factor (LIF) have been implicated in regulating astrogliosis following CNS lesions. Application of the factors activates astrocytes in vivo and in vitro, and their expression as well as their receptors is upregulated after brain injury. Here, we investigated their function by studying Müller cell activation induced by optic nerve crush in CNTF- and LIF-deficient mice, and in animals with deficiencies in cytokine signaling pathways. In the retina of CNTF(-/-) mice, basal GFAP expression was reduced, but unexpectedly, injury-induced upregulation in activated Müller cells was increased during the first 3 days after lesion as compared to wild-type animals and this corresponded with higher phosphorylation level of STAT3, an indicator of cytokine signaling. The observation that LIF expression was strongly upregulated in CNTF(-/-) mice but not in wild-type animals following optic nerve lesion provided a possible explanation. In fact, additional ablation of the LIF gene in CNTF/LIF double knockout mice almost completely abolished early lesion-induced GFAP upregulation in Müller cells and STAT3 phosphorylation. Early Müller cell activation was also eliminated in LIF(-/-) mice, despite normal CNTF levels, as well as in mutants deficient in gp130/JAK/STAT signaling and in conditional STAT3 knockout mice. Our results demonstrate that LIF signaling via the gp130/JAK/STAT3 pathway is required for the initiation of the astrogliosis-like reaction of retinal Müller cells after optic nerve injury. A potential role of CNTF was possibly masked by a compensatory increase in LIF signaling in the absence of CNTF.

Low-hemorrhage-risk surgical approach to expose the optic nerve in rabbits without bony removal and rectus resection

OBJECTIVE

A low-hemorrhage-risk surgical approach to expose the optic nerve (ON) in rabbits through the orbital process of the frontal bone without removal of the bony orbit and resection of the rectus muscle was explored and assessed in this study. This approach will be used to investigate a new visual prosthesis that requires intraorbital ON stimulation with penetrating electrodes. Animals Chinese Albino rabbits (n = 10).

METHODS

Rabbits were classified into a surgery and a control group (five in each). In the surgery group, the ON exposure was explored by the newly proposed surgical approach. Surgical time, blood loss, visually evoked potentials (VEP) at four different scheduled time points, and H&E-stained histology of the ON at one month after surgery were recorded and analyzed to assess the ease and safety of the approach.

RESULTS

The average surgical time for the ON exposure was 16.40 +/- 1.14 min with average blood loss of 0.52 +/- 0.08 mL. Within the one-month follow-up, the ON exhibited a naturally reversible conduction change in terms of VEP amplitude. Histological examination of the ON was unremarkable. A postoperative mild ptosis of the surgical eye resolved within one month after surgery.

CONCLUSION

The ease and safety of this new surgical approach allowed it to be easily used by non-expert operators and widely applied in rabbit experiments for various research purposes requiring exposure of the ON.

Morphological signs of apoptosis in axotomized ganglion cells of the rabbit retina

Optic nerve section in mammals induces apoptotic death of retinal ganglion cells (RGCs). However, a small population of RGCs survives for a relatively long time. These cells experience significant morphological changes due to the apoptotic process, but some of these changes are not clearly differentiated from those experienced in necrotic cells. In the present work, rabbit RGCs were studied 1 month after optic nerve section using light microscopy after neurobiotin injection, transmission electron microscopy (EM) and scanning electron microscopy (SEM). Apoptosis was identified by terminal deoxynucleotidyl transferase biotin-dUTP nick end labeling and characteristic signs of apoptosis were observed in the EM images. Ultrastructural analyses showed vacuolar degeneration in the cytoplasm and normal cellular structure loss. Signs of membrane changes were observed in axotomized RGCs by SEM. Early changes seen in the cell membrane suggest that axotomy may cause important changes in the cytoskeleton. We conclude that characteristic signs of apoptosis at the cell membrane level are clearly observed in rabbit RGCs after axotomy and they may be responsible for the cellular death.

Müller glial cells express nestin coupled with glial fibrillary acidic protein in experimentally induced glaucoma in the rat retina

This study was aimed to investigate reactive changes of Müller glial cells in rats subjected to experimentally induced glaucoma. In the latter, it is well documented that elevated intraocular pressure leads to the loss of ganglion cells as confirmed in this study. The present results have shown that Müller glial cells as well as astrocytes closely associated with the ganglion cells reacted vigorously to increased intraocular pressure as manifested by the induced and upregulated expression of nestin and glial fibrillar acidic protein. A major finding in glaucomatous rats was the induced expression of nestin together with glial fibrillar acidic protein with the rise of the intraocular pressure beginning at 2 h. The marked nestin expression appeared to be most intense at 1 week after operation and was sustained at 3 weeks. Induced nestin expression in Müller glial cells was demonstrated unequivocally in whole-mount preparation of the retina. In the same tissue preparation, nestin expression was also detected in some astrocytes. Western blotting analysis confirmed a marked increase in expression of nestin and glial fibrillar acidic protein. Present results suggest that nestin as well as glial fibrillar acidic protein is a useful biomarker for retina injury. The induced expression of these intermediate filament proteins in Müller glial cells especially at their end-feet and also in some astrocytes adjoining the neuronal injury suggests a potential neuroprotective mechanism in response to acute rise in intraocular pressure resulting in neuronal degeneration.

Effects of acute delivery of endothelin-1 on retinal ganglion cell loss in the rat

The vasoconstrictive peptide, Endothelin-1 (ET-1) has been found at elevated levels in glaucomatous eyes. In this study, a single 5mul intraocular injection of ET-1 was injected into the rat eye in order to characterize an in vivo retinal ganglion cell (RGC)-specific cell death model. The most effective concentration of ET-1 at inducing RGC loss at 2 weeks post-injection was determined using 5, 50 and 500mum concentrations of ET-1. The density of surviving RGCs was determined by counting Fluorogold labelled RGCs. A significant loss (25%) of RGCs was observed using only the 500mum concentration when compared to PBS-injected controls. GFAP immunohistochemistry revealed an increase in GFAP expression in Müller cell end-feet, as well as a total increase in GFAP expression (80%), following ET-1 treatment. These changes in GFAP expression are indicative of glial hyperactivity in response to stress. The specificity of ET-1 mediated cell death for RGCs was determined by measuring the changes in retinal thickness and TUNEL labeling. Retinal thickness was quantified using confocal and light microscopy. In confocal measurements, Yo Pro-1 was used to stain nuclear layers and the thickness of retinal layers determined from reconstructions. No significant loss in thickness was observed in any retinal layers. The same observations were seen in semi-thin sections when viewed by conventional transmitted light microscopy. The lack of significant thickness changes in the outer nuclear, outer plexiform or inner nuclear layer suggests that there was no significant cell loss in the retina other than in the RGC layer. Exclusive co-localization of TUNEL-labelled nuclei with Fluorogold-labelled cytoplasm provided additional evidence for RGC-specific death that most likely occurs via an apoptotic mechanism. A cell death time course was performed to determine RGC loss over time. RGC losses of 25, 25, 36 and 44% were observed at 1, 2, 3 and 4 weeks post-ET-1 injection, compared to PBS-injected controls. The total number of remaining RGC axons was determined by multiplying the number of optic nerve (ON) axons per unit area, by the cross-sectional area. There was a 31% loss in total ON axons in ET-1 treated eyes at 3 weeks post injection. Functional integrity of the visual system was determined by observing changes in the pupillary light reflex. ET-1 treatment resulted in a slowing of the pupil velocity by 31% and an average increase in the duration of contraction of 1.85sec (32% increase). These experiments provide evidence that acute ET-1 injections can produce RGC-specific cell death and many cellular changes that are similar to glaucoma. This potential glaucoma model leaves the optic nerve intact and may be used in subsequent experiments, which are involved in increasing RGC survival and functional recovery.

Microglial responses in the avascular quail retina following transection of the optic nerve

This study was undertaken to investigate microglial responses in the avascular central nervous system using the quail retina that is known to be devoid of blood vessels. Following intraorbital optic nerve transection (ONT), the quail retina was examined immunohistochemically at various times up to 6 months. A few days after transection, microglia in the inner retinal layers revealed features of activation. Activated cells displayed an amoeboid shape and enhanced QH1-immunoreactivity. The numbers of these amoeboid cells were rapidly increased, first in the inner plexiform layer (IPL), and then in the ganglion cell/nerve fiber layer (GCL/NFL) of the retina where retrograde degenerating ganglion cell processes and perikarya were located. By 6 months after transection, microglia regained their resting morphology, and their cell counts returned to control levels. At early time points of microglial activation, numerous QH1+ amoeboid cells were observed along the vitreal surface of the pecten and retinal region adjacent to the insertion of the pecten, where some amoeboid cells were attached underneath the internal limiting membrane, and appeared to squeeze through the optic nerve fiber bundles. A considerable number of these amoeboid cells in the GCL/NFL and the IPL were labeled with PCNA, suggesting that active exogenous migration (from the pecten) and in situ proliferation of precursor cells contribute to the increase in microglial population of the degenerating retina. On the other hand, TUNEL-positive microglia appeared in the GCL/NFL at later time points indicate that the decrease of microglial numbers is in part due to apoptosis in these layers. Although some aspects of microglial activation in the avascular retina appear unique, their consequences were similar to those described in vascular retinae of mammals, a finding indicates that blood vessels are not a prerequisite for microglial activation, and microglial precursors could migrate long distance to reach the lesioned site, which is not accessible via blood vessels. Our data provide the first analysis of microglial activation in the avascular central nervous system (CNS), and suggest that the quail retina is a useful model for studies of microglial behavior in CNS.

Expression of clusterin in Müller cells of the rat retina after pressure-induced ischemia

We have investigated the expression and cellular localization of clusterin in the rat retina following ischemia induced by transiently increasing the intraocular pressure. In the normal retina, weak clusterin immunoreactivity was visible in Müller cell profiles located in the inner nuclear layer. Following ischemia and reperfusion, strong immunoreactivity appeared in Müller cell somata and processes up to 3 days postlesion. Quantitative evaluation by immunoblotting confirmed that clusterin expression continuously increased and showed a peak value at 3 days after ischemic injury (to 1300% of control levels), and then decreased again to 400% of controls at 4 weeks postlesion. Immunocytochemistry using antisera against clusterin or glutamine synthase combined with the TUNEL method or immunocytochemistry using antisera activated caspase 3 and electron microscopy revealed that some clusterin-labeled Müller cells underwent apoptotic cell death. Our findings demonstrate that some Müller cells die by apoptosis, and suggest that clusterin produced and released by Müller cell may play an important role in the pathogenesis of ischemic injury in the rat retina.

Rabbit retinal ganglion cell survival after optic nerve section and its effect on the inner plexiform layer

Structural modifications of the inner retina were studied after optic nerve section (ONS) in the rabbit. Retinal ganglion cells (RGC) were labelled by injection of Fast Blue into the optic nerve, and counted under fluorescent light in control retina and retina 7, 14, 21 and 26 days post-axotomy. Studies on retinal cross-sections were also performed. For this purpose, retinal sections were stained with haematoxylin-eosin and immunohystochemistry for alpha1 and beta2/beta3 sub-units of the GABA(A) receptors. One week after axotomy, there was no significant loss in the number of ganglion cells with respect to control counts (1086+/-173cellsmm(-2) in the visual streak and 119+/-46cellsmm(-2) in the periphery, mean+/-SD, n=5). At 14 days post-axotomy, 271+/-46cellsmm(-2) remained in the visual streak and 33+/-6cellsmm(-2) in the periphery, corresponding to a mean survival of 27%. The number of ganglion cells decreased further on the following days, reaching 7.54% 1 month after ONS. A significant reduction in the thickness of the inner plexiform and ganglion cell layers was also observed in retinal cross-sections. Immunocytochemical studies show a remarkable disorganization of the layer stratification in the inner plexiform layer (IPL). We conclude that after ONS, RGC death occurs mainly between 7 and 14 days post-axotomy and a progressive death up to 26 days, causing a decrease in the thickness of the IPL and subsequent disorganization of its layers.

Morphometrical analysis of dendritic arborization in axotomized retinal ganglion cells

It has been reported that section of the optic nerve in mammals causes death in >90% of the retinal ganglion cells (RGCs). The cells which survive the section experience an irreparable loss of many of their dendritic segments and a rapid retraction of the dendritic tree. However, some growth cones and abnormal processes have been also reported. Our aim was to make a quantitative study of the morphological changes found in rabbit RGCs after optic nerve section. The morphometrical analysis of the RGCs which survived the axotomy showed an increase in the diameter of the soma and a significant increase in the area of the dendritic field; also, the length of the dendritic segments was significantly longer in axotomized RGCs than in control cells. Terminal dendritic segments (T) and preterminal segments (PT) were both measured in control and axotomized cells; the length ratio of T : PT segments was significantly greater in the axotomized cells than in the controls. We conclude that RGCs which survived the axotomy experienced a significant growth of their terminal dendritic branches.

Development of astrocytes in the lamina cribrosa sclerae of the mouse optic nerve, with special reference to myelin formation

In the mouse optic nerve, the optic nerve fiber layer in the retina, the optic papilla and the lamina cribrosa sclerae (LCS) just after penetrating the eyeball failed to generate myelin, whereas the optic nerve proper in the orbit was occupied by myelinated nerve fibers. The present study investigated development of the architecture of LCS, where the axons develop from unmyelinated to myelinated type, to elucidate how the initial part of axons was unmyelinated. At the LCS of the adult optic nerve, well developed astrocytes densely formed a cytoplasmic mesh-like frame through which unmyelinated fibers passed. The astrocytes here contained numerous and densely packed intermediate glial filaments and cell organelles. This framework formed by astrocytes appeared to be completed between 7 and 14 postnatal days before oligodendrocyte progenitors, migrated from the chiasm side, reached the proximal end of LCS, and began myelin formation. Thus the failure in myelin formation at the intraocular part and LCS possibly depended upon unsuccessful migration of oligodendrocytes beyond LCS constructed by specialized astrocytes, although other inhibitory factors for myelin formation, such as adhesion molecules distributed around LCS, may be unsolved.

Müller cells in allogeneic adult rabbit retinal transplants

Müller cell morphology and degree of activation in adult retinal transplants have, to our knowledge, never been reported previously. We transplanted adult rabbit neuroretinal full-thickness sheets, prepared under strict control, to the subretinal space of adult rabbits. After surviving 6-174 days, eyes were examined in the light microscope, and grafts displaying the normal laminated morphology were labeled with antibodies against vimentin and glial fibrillary acidic protein (GFAP). Müller cells in the grafts displayed the normal vertical arrangement, from outer limiting membrane to vitread endfeet. They showed an initial degree of activation, evident by GFAP upregulation, which diminished with increasing survival times, and was absent in the oldest specimens. In the host retina, Müller cells in the transplant area became progressively more disorganized with increasing survival times, and their degree of activation increased. Our results suggests that adult full-thickness neuroretinal grafts are structurally stable, even in long-term specimens, and thrive in spite of their allogeneic environment. The gliotic change seen in the host retina covering the graft is identical to the one seen in earlier reported eyes receiving embryonic grafts, and is due to the merangiotic nature of the rabbit neuroretina.

Down-regulation of glutamate-induced conductances of retinal horizontal cells after ganglion cell axotomy

After a complete optic nerve section (ONS), retinal neurons may display retrograde transneuronal modifications in synaptic structure and function related to the retinal disconnection from the brain. The molecular and physiological basis of these changes is not yet fully understood. Immunoreactivity for calbindin was used to specifically immunolabel the horizontal cells (HC) in order to study any morphologic changes in the outer plexiform layer (OPL) after axotomy-induced degeneration of retinal ganglion cells (RGC) in the rabbit retina. Glutamate-gated conductance expressed by HC enzymatically dissociated from the rabbit retina were studied at 12 and 21 days after ONS by using the whole-cell voltage-clamp technique. The amplitudes of glutamate-induced currents on HC were significantly reduced 3 weeks after axotomy. However, no morphologic changes within the OPL were detected coincident with the progressive loss of glutamatergic responses; similarly, HC dissociated from the axotomized retinal tissue did not differ in morphology or appearance from control retinas. The main finding in this study is that the HC experiment a retrograde transneuronal down-regulation of their ionotropic glutamate-induced conductance following axotomy-induced degeneration of RGC.

Expression of glial fibrillary acidic protein and glutamine synthetase by Müller cells after optic nerve damage and intravitreal application of brain-derived neurotrophic factor

Müller glia play an important role in maintaining retinal homeostasis, and brain-derived neurotrophic factor (BDNF) has proven to be an effective retinal ganglion cell (RGC) neuroprotectant following optic nerve injury. The goal of these studies was to investigate the relation between optic nerve injury and Müller cell activation, and to determine the extent to which BDNF affects the injury response of Müller cells. Using immunocytochemistry and Western blot analysis, temporal changes in the expression of glial fibrillary acidic protein (GFAP) and glutamine synthetase (GS) were examined in rats after optic nerve crush alone, or in conjunction with an intravitreal injection of BDNF (5 microg). GFAP protein levels were normal at 1 day post-crush, but increased approximately 9-fold by day 3 and remained elevated over the 2-week period studied. Müller cell GS expression remained stable after optic nerve crush, but the protein showed a transient shift in its cellular distribution; during the initial 24-h period post-crush the GS protein appeared to translocate from the cell body to the inner and outer glial processes, and particularly to the basal endfeet located in the ganglion cell layer. BDNF alone, or in combination with optic nerve crush, did not have a significant effect on the expression of either GFAP or GS compared with the normal retina, or after optic nerve crush alone, respectively. The data indicate that although BDNF is a potent neuroprotectant in the vertebrate retina, it does not appear to have a significant influence on Müller cell expression of either GS or GFAP in response to optic nerve injury.

Müller cells in long-term full-thickness retinal transplants

Müller cells are essential in creating and maintaining intricate neuroretinal architecture. The functions of this important glial cell are not limited to mere support of the retinal neurons, but also include interaction in synaptic transmission and activation in response to retinal insult. In this study, we have examined Müller cell morphology and degree of activation in embryonic full-thickness rabbit neuroretinal grafts, which were positioned under the host retina using vitrectomy technique. After surviving 3-10 months, retinal specimens were examined with hematoxylin and eosin staining and immunohistochemical analysis of vimentin and glial fibrillary acidic protein (GFAP) expression. In the host retina covering the graft, outer layers were degenerated, and vimentin-labeled Müller cells in this area appeared short, disorganized, and displayed strong GFAP labeling. In the graft, vimentin-labeled Müller cells spanning the retinal layers in the normal manner were found. Müller cells in 3-month grafts were well labeled by GFAP, whereas in older grafts, GFAP labeling was very weak or absent. Our results suggest that Müller cells in well-laminated full-thickness retinal grafts display many of the normal morphological features and retain a normal organization even after prolonged survival times. The loss of the initial degree of Müller cell activation indicates a long-term stability of the graft. The degeneration and gliosis of the host retina covering the graft is best explained by the merangiotic nature of the rabbit retina and may limit the usefulness of the rabbit in retinal transplantation experiments.

A different retinal glia response to optic nerve injury/lipopolysaccharide administration in hooded and albino rats

Despite a massive degeneration of retinal ganglion cells (RGC) after optic nerve crush (ONC) in hooded rats only a minor increase in retinal glial fibrillary acidic protein (GFAP)-immunoreactivity was found in the inner retina. Interestingly, a combination of ONC with the administration of the proinflammatory agent lipopolysaccharide (LPS) but not LPS alone induces increased GFAP-immunoreactivity. In contrast albino rats showed elevated GFAP-immunoreactivity in response to both, LPS-administration and ONC with no further increase after a combination of both. These data demonstrate significant differences in retinal glia responsiveness between hooded and albino rats after optic nerve lesions.

Expression of brain/kidney protein in Müller cells of rat retina following transient ischemia

We have investigated the expression and cellular localization of brain/kidney (B/K) protein in the rat retina following transient ischemia. In the normal retina, strong B/K immunoreactivity was localized to some ganglion cells. In addition, a few radial Muller cell processes showed B/K immunoreactivity. Following ischemia and reperfusion, most B/K-labeled ganglion cells were lost, whereas between 1 day and 2 weeks post-lesion B/K immunoreactivity appeared in many more Muller cell processes with increasing intensity. Quantitative evaluation by immunoblotting confirmed that B/K expression then decreased progressively, to 35% of control values at four weeks post-lesion. Our findings suggest that Muller cells are involved in the pathophysiology of retinal ischemia through the expression of B/K following transient ischemia.

Glial cells in the bird retina: immunochemical detection

The avian retina is remarkably different from its mammalian counterpart in macroglial cell appearance. First, it is completely devoid of astrocytes. Thus, Müller cells constitute the only astrocytic-like cell population in avian retinae, whereas mammalian retinae also contain astrocytes in close association with blood vessels. Second, axons in the optic nerve layer of the retina of birds are myelinated, unlike those found in most mammalian species, with the exception of the rabbit, in which the medullary rays of the retina are myelinated by oligodendrocytes. Recent studies have revealed evidence that bird retinae contain a large number of oligodendrocytes, but which glial cell type myelinates axons intraretinally is still controversial. Apart from macroglial appearance, microglia in the bird retina show a very similar pattern of distribution to that of mammalian counterparts. This article reviews the existing data, including our new observations, and discusses the issues that remain to be resolved.

Upregulation of ciliary neurotrophic factor in reactive Müller cells in the rat retina following optic nerve transection

We have investigated the expression and cellular localization of ciliary neurotrophic factor (CNTF) in the rat retina following optic nerve transection. In the normal retina, CNTF immunoreactivity was restricted to profiles in the ganglion cell layer. Following optic nerve transection, immunoreactivity appeared in Müller cell somata and processes and its intensity increased between three and seven days post-lesion. Quantitative evaluation by immunoblotting confirmed that CNTF expression increased continuously up to 7 days after optic nerve transection (to 430% of control levels), but decreased again to 250% of controls at 4 weeks post-lesion. Our findings suggest that CNTF supplied by Müller cells may play a protective role for axotomized ganglion cells in the rat retina.

Localization of IgG in the normal and dystrophic rat retina after laser lesions

PURPOSE

To test the hypothesis that access to extravasated plasma protein IgG may influence photoreceptor survival following laser photocoagulation and to determine whether this correlates with the retinal glial reaction.

METHODS

A total of 45 rats (18 Royal College of Surgeons (RCS) dystrophic and 18 RCS-rdy+ congenic control) were used for this experiment. Nine non-lasered littermates of same age were used as controls. The superior retinas of postnatal day 23 rats were irradiated with a grid pattern of 40 argon green laser lesions of 50 microm in diameter and two powers (150 and 300 mW) for 0.2 s. At various times after laser lesions (up to 14 days), animals were perfused, the retinas snap frozen and sectioned on a cryostat. A one-step immunohistochemical technique was used by incubating with rabbit anti-rat IgG conjugated directly to horseradish peroxidase. Adjacent sections were processed using an antibody to glial fibrillary acidic protein (GFAP) by the standard avidin-biotin complex method.

RESULTS

The labelling pattern for extravasated IgG after laser lesion was very similar in both RCS and RCS-rdy+ rat retinas. At 6, 12 and 24 h after lesions, IgG immunoreactivity (IR) was very intense in the lesion core and flanks. The outer plexiform layer (OPL) and photoreceptor inner segments provided a ready pathway for lateral spread of IgG. However, in the outer nuclear layer (ONL), IgG localization was much more restricted. Despite very intense IgG IR in the ONL of the coagulated lesion core, there was always a very sharply delineated boundary where the label abruptly halted. The GFAP labelling in both RCS dystrophic and RCS-rdy+ congenic control rat retinas showed that this boundary was between normal and necrotic cells because there was a core where GFAP was not produced by Müller cells. By 2 days after lesions, the coagulated cells in the lesion core were being removed by phagocytic cells that were IgG IR. Labelled phagocytic cells were also found among the inner and outer segment region on the lesion flanks. There was still IgG IR in the lesion, but the label was faint. No IgG IR was found in the retina at 3, 4, 7 and 14 days after lesions. Absorption control with pure rat IgG showed the label to be specific.

CONCLUSIONS

The extravasated IgG was derived from the choroidal circulation because at no stage was IgG localized around the retinal vasculature. The IgG labelling was surprisingly widespread and, therefore, did not correlate with photoreceptor sparing, although it preceded the widespread Müller cell expression of GFAP and may, therefore, trigger glial reaction.

Expression of CNTF in Müller cells of the rat retina after pressure-induced ischemia

We have investigated the expression and cellular localization of ciliary neurotrophic factor (CNTF) in the rat retina following ischemia induced by transiently increasing the intraocular pressure. In the normal retina, CNTF immunoreactivity was restricted to profiles in the ganglion cell layer. Following ischemia and reperfusion, immunoreactivity appeared in Müller cell somata and processes and its intensity increased between 1 day and 2 weeks post-lesion. Quantitative evaluation by immunoblotting confirmed that CNTF expression continuously increased up to 2 weeks after ischemic injury (to 600% of control levels), but had declined again to 250% of controls at 4 weeks post-lesion. Our findings suggest that CNTF supplied by Müller cells has a protective function for lesioned neurons following transient ischemia.

Regeneration in the developing optic nerve: correlating observations in the opossum to other mammalian systems

Regeneration of severed axons within the central nervous system of adult mammals does not normally occur with any degree of success. During development, however, newly forming projections must send axons to distant sites and form appropriate connections with their targets: successful regeneration has been observed during this critical period. The opossum central nervous system develops during early postnatal life and has provided a useful experimental model to investigate this specialized mode of axonal regeneration in mammals. The presence of a clear decision point at the optic chiasm has also provided a useful site at which to investigate the navigational capacity of retinal ganglion cells regenerating along the optic nerve during this critical period. Regeneration failure occurs as the central nervous system progresses from this permissive, developing state to a mature, non-permissive adult state. Studies into the behaviour of glial and neuronal elements around this transition period can help elucidate some of the factors that need to be overcome if regeneration is ever to become successful in adult mammals. The regeneration characteristics of a lesioned projection are dependent upon its developmental stage and are also related to the proximity of axotomy along its pathway. A system of staging is proposed to correlate observations in the opossum optic nerve to other mammalian systems.

Expression of myelin proteins in the opossum optic nerve: late appearance of inhibitors implicates an earlier non-myelin factor in preventing ganglion cell regeneration

The pattern of appearance of myelin-associated proteins in the visual system of the Brazilian opossum Monodelphis domestica is described. Whole mounts of optic nerve, chiasm, and optic tract were sectioned horizontally and incubated with antibodies to myelin basic protein (MBP), proteolipid protein (PLP), myelin-associated glycoprotein (MAG), "Rip," and the neurite inhibitory protein (IN-1), followed by visualization with diaminobenzidine and a peroxidase-conjugated secondary antibody. PLP is first detectable 24 days after birth (P24) at the centre of the optic chiasm. MBP, MAG, Rip, and IN-1 appear first in the same area at P26. By P28 the distribution of all proteins is similar, occupying the entire chiasm, optic tracts, and prechiasmatic portion of the optic nerves. Protein expression progresses along the optic nerve to reach the lamina cribrosa by P34, coincident with the time of eye opening. A critical period in which the retinofugal pathway has a regenerative capacity has recently been observed in Monodelphis. This period ends at P12, 2 weeks before the appearance of the myelin-associated inhibitory proteins MAG and IN-1. These results therefore suggest that regeneration in the developing retinofugal projection of the opossum is restricted by an earlier non-myelin factor, which is in contrast to current literature on the spinal cord.

Development and role of retinal glia in regeneration of ganglion cells following retinal injury

AIMS/BACKGROUND

Recent observations have shown that the glial scar resulting from a surgical lesion of the immature retina differs from elsewhere in the central nervous system, in that it permits the through growth and reconnection of regenerating axons. This study in the opossum examines in detail the development and reaction to injury of retinal glia at different developmental stages, and specifically examines the distribution of the gliosis related inhibitory molecule, chondroitin sulphate proteoglycan (CSPG), making comparisons with a control site of gliosis in the cerebral cortex.

METHODS

A linear slit was cut into the retina or cortex with a fine tungsten probe. After a variable time delay, immunocytochemistry of the resulting gliosis was employed to detect astrocytes with glial fibrillary acidic protein (GFAP), Müller cells with vimentin, and CSPG with CS-56 antibodies. GFAP was also used at different ages to examine the normal development of astrocytes in the retina of this species.

RESULTS

Astrocytes entered the retina 12 days after birth (P12), closely associated with blood vessels in the nerve fibre layer. In experiments at all ages studied, cellular continuity was re-established across the lesioned retina, which did not result in a significant astrocyte proliferation or CSPG expression. In contrast, cortical injury led to the development of a cystic cavity surrounded by astrocytes and CSPG. Müller cells expressed GFAP but not CSPG in the lesioned retina.

CONCLUSION

Successful regrowth of ganglion cells through a retinal lesion may be partly the result of the scarcity of astrocytes in the retina, which results in minimal gliosis, or of their apparent inability to express inhibitory molecules.

Microglial responses to focal lesions of the rabbit retina: correlation with neural and macroglial reactions

There is a very wide spread Müller glial response to focal laser photocoagulation lesions in the rabbit retina. In this study we have described the microglial response to similar lesions and compared this with the Müller and retinal ganglion cell responses. Microglia were labelled using nucleoside di-phosphatase histochemistry in adult rabbit retinal wholemounts and compared with axonal and Müller cell responses as shown respectively by neurofilament and GFAP immunohistochemistry. In the normal retina, microglia were located in the nerve fibre layer (NFL), inner plexiform layer (IPL), and sparsely in the outer plexiform layer (OPL). Following laser photocoagulation each layer reacted differently. The NFL reaction was exclusively associated with axonal degeneration, as shown by abnormal neurofilament label, and therefore only started several days after injury. In the IPL, neighbouring microglial cells directed their processes towards the lesion by 2 h and had migrated into the lesion by 6 h, but the reaction did not extend more than 2-3 cell diameters from the lesion and was over by 7 days. In the OPL the cell density increased by 1-2 days over a few millimeters from the lesion. The Müller cells expressed GFAP for several millimeters from the lesion starting at 24 h and persisting for over one month and therefore the correlation with the microglial reaction was poor. The different reaction in each retinal layer is evidence that microglial responses are modulated by local factors, probably mainly by contact with injured retinal elements as well as diffusable factors.

Expression of the axonal cell adhesion molecules axonin-1 and Ng-CAM during the development of the chick retinotectal system

Cell surface glycoproteins expressed on growth cones and axons during brain development have been postulated to be involved in the cell-cell interactions that guide axons into their target area. Nevertheless, an unequivocal description of the mechanism by which such molecules exert control over the pathway of a growing axon has not been done. As a crucial requirement in support of a relevant involvement of an axonal surface molecule in growth cone guidance, this molecule should be expressed in the growth cone. The developing retinotectal system provides an excellent opportunity to test whether a particular neuronal surface molecule fulfills the requirement of the spatiotemporal coincidence between its appearance and the emergence of growth cones because its setup follows the rule of chronotopy, i.e., the position of axons in a certain site is determined by the time of their arrival. We have analyzed axonin-1 and the neuron-glia cell adhesion molecule (Ng-CAM), two axonal surface molecules that promote neurite growth in vitro, for their expression in the retina and in the retinotectal system of the chick throughout its development. At stage 18, both axonin-like (A-LI) and Ng-CAM-like immunoreactivity (Ng-CAM-LI) are clearly present in the area where first retinal ganglion cells (RGCs) are generated. The immunoreactivity spreads synchronously with the formation of RGCs over the developing retina. From stage 32 on, the inner plexiform layer is also stained according to its temporospatial gradient of maturation. In later stages, the outer plexiform layer and the inner segments of photoreceptors also show immunoreactivity. The development of A-LI and Ng-CAM-LI along the optic nerve, chiasm, optic tract, and in the superficial layers of the optic tectum follows the chronotopic pattern of axons, as was found by earlier morphological investigations. Older axons loose their A-LI. This allows to localize the position of newly formed axons. The fact that A-LI and Ng-CAM-LI parallel the formation and maturation of axons suggests that axonin-1 and Ng-CAM may play an important role in the organization of the retinotectal system.

NADPH diaphorase expression in the rat retina after axotomy--a supportive role for nitric oxide

The large majority of mammalian retinal ganglion cells degenerate following section of their axons in the optic nerve. It has been suggested that some axotomized retina ganglion cells die because of toxic agents produced within their immediate environment. Our hypothesis was that nitric oxide might be one of the toxic factors implicated in the death of adult retinal ganglion cells post-axotomy. In the first instance, we determined whether there were any changes in the retinal expression of NADPH diaphorase both 3 and 14 days following intraorbital section of the optic nerve in adult rats. Secondly, if nitric oxide was indeed implicated in the death of ganglion cells, then trophic factors which rescue these neurons might do so by decreasing the expression of nitric oxide synthase. Recently, we found that a collicular proteoglycan purified from the major target of retinal ganglion cells, the superior colliculus, rescued a greater proportion of adult ganglion cells from axotomy-induced death than most other known trophic factors. We thus injected this proteoglycan intraocularly after section of the optic nerve and examined its effect on the expression of NADPH diaphorase in the retina. Thirdly, an inhibitor of nitric oxide synthetase was repeatedly injected into the eye following the section of the optic nerve in order to determine if such a treatment might improve the survival of retinal ganglion cells. The present results indicate that section of the optic nerve does not alter the overall levels of NADPH diaphorase within the adult rat retina. Intraocular injections of the collicular proteoglycan actually increased the number of neurons expressing NADPH diaphorase, particularly in the ganglion cell layer. Finally, inhibition of nitric oxide synthetase following axotomy resulted in increased loss of retinal ganglion cells over a 2 week period when compared with controls. Our findings indicate that, rather than being toxic, small amounts of nitric oxide may be important for the survival of a proportion of injured retina ganglion cells.

Glial reactivity in the retina of adult rats

The present study aimed to characterize the reaction of mammalian (rat) retinal macroglia (Müller cells and astrocytes) to disturbances of their environment in the form of intraorbital section of the optic nerve, intraocular insertion of a thin glass capillary (without damage to the retina) or a combination of both. Glial reactivity was assessed through the use of a battery of antibodies which recognise four different proteins--glial fibrillary protein (GFAP) and three other proteins designated respectively MA1, 4D6 and 4H11. Retinal astrocytes did not exhibit any changes in normally expressed GFAP or MA1. By contrast, the expression of GFAP and MA1 in Müller cells increased 14 days following section of the optic nerve and/or intravitreal insertions of a glass capillary. Three days postoperatively, the expression of GFAP, but not MA1, had already increased significantly in Müller cells. 4D6 and 4H11 proteins were not expressed in astrocytes. In Müller cells, the levels of these proteins increased significantly following combined optic nerve section and intraocular insertion of a glass capillary. Thus, a mechanical disturbance of the intraocular environment constitutes a more effective stimulus in increasing the expression of some Müllerian proteins than damage to the axons of retinal ganglion cells. Such changes have important implications for various ocular treatments that involve intraocular administration of drugs, as well as for the survival/regeneration potential of retinal ganglion cells undergoing Wallerian degeneration.

A critical period for axon regrowth through a lesion in the developing mammalian retina

Although the central nervous system of mature mammals is incapable of regeneration, certain elements present in the developing system must permit and promote the growth of new axons to their initial targets. We investigate whether the environment of a developing visual system is capable of supporting regeneration in the Brazilian opossum Monodelphis domestica, in which the retinofugal system develops postnatally. Retinae were lesioned up to the 16th postnatal day and analysed for regeneration after a further 7-10 days. Anterograde tracing with Dil showed axons to have regrown from the axotomized area of retina directly through the lesion. Retrograde tracing with horseradish peroxidase injected into the superior colliculus confirmed that axons from the lesioned area of retina had grown to an appropriate position in the midbrain. The proportion of retinae in which axonal continuity was restored across the lesion decreased as the visual system matured, falling to zero after the 12th postnatal day. Thus a critical period exists in the postnatal opossum in which a retinal lesion permits axon passage. Correlating these results to the known pattern of retinofugal pathway development provides an insight into factors that may restrict this critical period to the 12th postnatal day, and suggests that at least some of the axotomized neurons are regenerating.

Number and dendritic morphology of retinal ganglion cells that survived after axotomy in adult cats

Retinal ganglion cells (RGCs) of adult cats were labeled by injection of diI into the proximal stump of completely transected optic nerves. Approximately 2% to 5% of the RGC population appeared viable 2 months after these axotomies, based on diI retention. The morphological type and dendritic arbor of these surviving RGCs were examined after intracellular injections of Lucifer Yellow into diI-labeled RGCs. Postaxotomy survival rate was much higher for alpha-like cells than for beta-like cells. However, in one of four retinas examined, a large number of RGCs seemed to survive axotomy, and among these, beta cells survived at an unusually high rate. Dendritic arbors of surviving RGCs were also examined after intracellular injection of horseradish peroxidase. Some dendrites of these RGCs lacked branches and were thin in caliber. Other dendrites displayed many spiny processes and bulbous swellings. Essentially, these results confirm the previous suggestion that alpha cells survive axotomy longer than beta cells. The ability of alpha cells to regenerate axons may thus be attributable to their relatively high resistance to axotomy. The atypical dendritic profiles seen after optic nerve transection may reflect either degeneration or regrowth of dendrites.

Differential vulnerability of neurochemically identified subpopulations of retinal neurons in a monkey model of glaucoma

The vulnerability of subpopulations of retinal neurons delineated by their content of cytoskeletal or calcium-binding proteins was evaluated in the retinas of cynomolgus monkeys in which glaucoma was produced with an argon laser. We quantitatively compared the number of neurons containing either neurofilament (NF) protein, parvalbumin, calbindin or calretinin immunoreactivity in central and peripheral portions of the nasal and temporal quadrants of the retina from glaucomatous and fellow non-glaucomatous eyes. There was no significant difference between the proportion of amacrine, horizontal and bipolar cells labeled with antibodies to the calcium-binding proteins comparing the two eyes. NF triplet immunoreactivity was present in a subpopulation of retinal ganglion cells, many of which, but not all, likely correspond to large ganglion cells that subserve the magnocellular visual pathway. Loss of NF protein-containing retinal ganglion cells was widespread throughout the central (59-77% loss) and peripheral (96-97%) nasal and temporal quadrants and was associated with the loss of NF-immunoreactive optic nerve fibers in the glaucomatous eyes. Comparison of counts of NF-immunoreactive neurons with total cell loss evaluated by Nissl staining indicated that NF protein-immunoreactive cells represent a large proportion of the cells that degenerate in the glaucomatous eyes, particularly in the peripheral regions of the retina. Such data may be useful in determining the cellular basis for sensitivity to this pathologic process and may also be helpful in the design of diagnostic tests that may be sensitive to the loss of the subset of NF-immunoreactive ganglion cells.

Receptive-field properties of displaced starburst amacrine cells change following axotomy-induced degeneration of ganglion cells

Starburst amacrine cells in the rabbit retina were labeled following an intraocular injection of the fluorescent dye, 4,6,diamidino-2-phenylindole (DAPI). From each eye a strip of retina was removed, mounted on a platform beneath an epifluorescence microscope, and superfused with a physiological solution. The tip of a tungsten microelectrode (for extracellular recording) was visually positioned near the cell body of a DAPI-labeled starburst amacrine cell that was located in the ganglion cell layer. Light-evoked responses from the displaced starburst amacrine cells were studied in normal retinas and in retinas that had received a small electrolytic lesion near the optic disk 5-9 months beforehand. In normal retinas, a small spot of light centered over the receptive field of a displaced starburst amacrine cell in nearly all cases evoked a brief burst of spikes only at light onset. When stimulated with a large spot or an annulus of light, many cells gave a small burst of spikes at light offset. In lesioned retinas, the light responses of displaced starburst amacrine cells were recorded in areas of the retina where ganglion cells had degenerated. All cells responded with a large burst of spikes at the onset and offset of a small, centered spot of light. Large spots and annuli of light also evoked robust ON/OFF responses from these cells. The results from this study show that the receptive-field properties of displaced starburst amacrine cells change following axotomy-induced degeneration of ganglion cells. This finding indicates that changes in either synaptic transmission or the membrane properties of neurons occur in the retina following degeneration of ganglion cells.

Neurotoxic effects of neonatal injections of monosodium L-glutamate (L-MSG) on the retinal ganglion cell layer of the golden hamster: anatomical and functional consequences on the circadian system

In rodents, daily injection of neurotoxic monosodium L-glutamate (MSG) during the postnatal period induces retinal lesions, optic nerve degeneration with an alteration of visual pathway and an absence of the b-wave in the electroretinogram. Despite this damage, electrophysiological responses subsist in the lateral geniculate bodies and synchronization of circadian rhythms to the light/dark cycle can still occur. Using two formal properties of the circadian system (entrainment and phase-shift by light), we assessed the functionality of retinal projections to the circadian clock in MSG-treated hamsters. Displaced amacrine and ganglion cell populations were quantified and retinal terminals in the suprachiasmatic nuclei were estimated. Animals received daily doses of glutamate during the first ten days after birth according to two protocols. The two treatments similarly destroyed 56% of the overall population of the ganglion cell layer: 30% of displaced amacrine and 89% of ganglion cells. Surviving ganglion neurons (7,500 cells) were evenly distributed across the entire retina except in one area of high cell density located in the temporoventral quadrant. Retinal projections of the "image-forming" pathway were drastically reduced in the dorsal lateral geniculate bodies, less in their ventral part. The "nonimage-forming" pathway was also affected since the volume of labeled terminals in the suprachiasmatic nuclei was reduced by one-half to one-third. Nevertheless, treated hamsters exhibited a free-running locomotor activity rhythm after several months in constant darkness, could be entrained by the light/dark cycle and phase-shifted by light pulses. These results suggest that a damaged retinohypothalamic tract can still assume the photic entrainment of the circadian clock.

Unidirectional coupling of gap junctions between neuroglia

Gap junctions permit the passage of ions and small molecules between cells, thereby providing a basis for direct intercellular communication. In the rabbit retina, the low molecular weight dyes Lucifer yellow and biocytin passed readily from astrocytes into adjacent astrocytes, oligodendrocytes, and Müller cells. However, the dyes rarely passed from either oligodendrocytes or Müller cells into astrocytes. Unidirectional passage of dye suggests the presence of an asymmetric barrier to the movement of molecules through heterologous gap junctions and indicates the potential for a hierarchy of command between interconnected cells.

Neuropeptide Y immunoreactivity identifies a group of gamma-type retinal ganglion cells in the cat

Ganglion cells within the cat retina have been traditionally grouped by morphological criteria into three major classes: alpha, beta, and gamma. The gamma-type cells have been least well characterized, but the available evidence indicates that this class comprises a relatively heterogeneous population of neurons. In the present study we demonstrate that an antibody for neuropeptide Y (NPY) recognizes a subpopulation of about 2,000 gamma-type ganglion cells. The NPY-immunoreactive (IR) neurons project to the superior colliculus and to the C layers of the lateral geniculate nucleus as demonstrated by retrograde labeling with fluorescent tracers (fluorogold or rhodamine latex microspheres). Virtually all of these cells disappear following lesions of the optic nerve. The NPY-IR ganglion cells were identified as gamma cells on the basis of soma size and dendritic branching patterns. The somas of these neurons are small (8-22 microns in diameter), and each cell is characterized by sparsely branching dendritic processes, usually extending into the middle third of the inner plexiform layer, the physiologically defined ON sublamina. These neurons are distributed across the entire retina, with the highest density at the area centralis. Within local regions of the retina, however, there was no indication that the NPY-IR gamma cells are arrayed in a regular mosaic pattern. These results provide the first evidence that the gamma class of ganglion cells of the cat retina can be subdivided on the basis of immunocytochemical properties.

Muller cell expression of glial fibrillary acidic protein (GFAP) in RPE-cell transplanted retinas of RCS dystrophic rats

In Royal College of Surgeons (RCS) rats with inherited retinal dystrophy, photoreceptor cell degeneration is accompanied by Muller cell changes, in particular, increased expression of the intermediate filament protein, glial fibrillary acidic protein (GFAP). In this study, we examined the expression of GFAP in retinas of 4 month-old RCS dystrophic rats either transplanted with normal retinal pigment epithelial (RPE) cells or injected with vehicle (sham control). The sham-injected and non-treated retinas showed increased expression of GFAP in Muller cells with advancing age. GFAP-immunostained Muller processes were observed in the region of the subretinal space of these RCS retinas beginning by 4 months. However, in the retinas of RCS rats transplanted with normal RPE cells, GFAP expression by Muller cells was significantly reduced. Specifically, under the transplanted RPE cells, GFAP-immunolabeled Muller radial processes were not observed and GFAP immunostaining was not present in the subretinal space at any time period examined. Immunoblot analysis of the superior hemisphere of RPE-cell transplanted retinas of 4 month-old RCS dystrophic rats showed less GFAP immunostaining than sham-injected and the inferior hemisphere of retinas of age-matched RCS rats. This study showed that in normal RPE-cell transplanted retinas of RCS rats, in addition to rescued photoreceptor cells, there was an accompanying stabilization of Muller cells as demonstrated by a reduction in the expression of GFAP.

A quantitative study of the lateral spread of Müller cell responses to retinal lesions in the rabbit

A wide variety of retinal pathology is associated with an increase in Müller glial cell expression of glial fibrillary acidic protein (GFAP). In this study the time course and spatial spread of the Müller cell GFAP response following argon laser photocoagulation lesions was examined in wholemounted rabbit retina. At 24 hours single focal lesions were surrounded by GFAP positive Müller cell end feet which declined in density with distance but extended as far as 2-3 mm from the lesion. The Müller cell reaction reached a maximal spread of 4-5 mm at 14 to 21 days and had started to contract by 30 days, leaving a core of GFAP positive processes immediately around the lesion site at 60 days. This zone of spread was much larger than the area of disrupted pigment epithelium. Isodensity plots did not reveal any correlation with the trajectory of retinal ganglion cell axons. The spread of reaction was more confined for lesions within the visual streak than in the dorsal or ventral retinal periphery. Multiple lesions within a focal region of retina resulted in a greater density of GFAP reactive end feet with a corresponding greater spread. However, when five to ten lesions were made in a horizontal row, the Müller cells over the entire retina became GFAP immunoreactive. This pan-retinal reaction took several days to spread, peaked at 7-14 days, and contracted back to the primary lesion sites by 2 months. This spread of Müller cell reactivity may be triggered by the diffusion of substances released by injury or it may be due to direct cellular communication. The extensive indirect effect on Müller cells of laser irradiation might be an important component of the clinical effect of laser photocoagulation and indicates a long distance communication mechanism between retinal glia which is poorly understood. This study also shows the importance of the time at which the Müller cell response is assessed.

Long-term induction of c-jun mRNA and Jun protein in rabbit retinal ganglion cells following axotomy or colchicine treatment

The expression of the c-jun, c-fos, and NGFI-A genes was studied in the rabbit retina after optic nerve crush (ONC) or an intravitreal injection of colchicine. By Northern blotting, the basal expression of c-fos and NGFI-A mRNAs were undetectable, whereas c-jun mRNA showed a low basal expression in sham-operated control retinas. Very few or no Jun- or Fos-immunoreactive nuclei were seen in control retinas. From 1 to 95 days after ONC a marked induction of JUN- but not FOS-immunoreactive neurons was seen in the ganglion cell layer peaking at 3 and 7 days. Jun-positive neurons also accumulated immunoreactive phosphorylated neurofilaments, indicating that they were ganglion cells. Northern blots demonstrated that retinal levels of c-jun mRNA, but not of c-fos or NGFI-A mRNAs, were increased 3 and 7 days after ONC. An intravitreal injection of colchicine also induced Jun-immunoreactivity within 24 hr in most of the neurons in the ganglion cell layer, but not in the inner nuclear and outer nuclear layers. The results indicate that axonal damage induces a specific pattern of IEG expression including a long-term induction of the c-jun gene in CNS neurons.