Pax2 in the optic nerve of the goldfish, a model of continuous growth

Zebrafish embryonically exposed to valproic acid present impaired retinal development and sleep behavior

Prenatal exposure to valproic acid (VPA), a drug widely used to treat epilepsy and bipolar disorder, is an environmental risk factor for autism spectrum disorder (ASD). VPA has been used to reproduce the core symptoms of ASD in animal model organisms, including zebrafish. Visual system functioning is essential in the interpretation of social conditions and plays an important role of several behavioral responses. We hypothesized that behavioral deficits displayed by ASD patients may involve impaired visual processing. We used zebrafish as model organism to investigate the visual system after embryonic exposure to VPA using histological, behavioral and gene expression analysis. We analyzed the pineal gland of zebrafish and sleep-like behavior to study how VPA exposure alters photo-sensibility of zebrafish. VPA-exposed zebrafish showed a delay in the development of the retina and optic nerve, which normalized at five days post fertilization. At larval stage, VPA-exposed zebrafish showed sleep disturbances associated with a reduced number of serotonin-producing cells of the pineal gland. In addition, the number of hypocretin/orexin (hcrt) expressing neurons in the rostral hypothalamus at 6 and 14 days post fertilization was reduced. In conclusion, we demonstrated that although VPA exposure leads to a delay in visual system development, it does not affect larval visual function. The novel finding that VPA alters significantly cells involved in sleep regulation and the sleep-like state itself may be relevant for understanding sleep disturbances in ASD patients.

Doublecortin in the Fish Visual System, a Specific Protein of Maturing Neurons

Simple Summary

Doublecortin (DCX) is an essential protein in the development of the central nervous system and in lamination of the mammalian cortex. It is known that the expression of DCX is restricted to newborn neurons. The visual system of teleost fish has been postulated as an ideal model since it continuously grows throughout the animal’s life. Here, we report a comparative expression analysis of DCX between two teleost fish species as well as a bioinformatic analysis with other animal groups. Our results demonstrate that DCX is very useful for identifying new neurons in the visual systems of Astatotilapia burtoni, but is absent in Danio rerio.

Abstract

Doublecortin (DCX) is a microtubule associated protein, essential for correct central nervous system development and lamination in the mammalian cortex. It has been demonstrated to be expressed in developing—but not in mature—neurons. The teleost visual system is an ideal model to study mechanisms of adult neurogenesis due to its continuous life-long growth. Here, we report immunohistochemical, in silico, and western blot analysis to detect the DCX protein in the visual system of teleost fish. We clearly determined the expression of DCX in newly generated cells in the retina of the cichlid fish Astatotilapia burtoni, but not in the cyprinid fish Danio rerio. Here, we show that DCX is not associated with migrating cells but could be related to axonal growth. This work brings to light the high conservation of DCX sequences between different evolutionary groups, which make it an ideal marker for maturing neurons in various species. The results from different techniques corroborate the absence of DCX expression in zebrafish. In A. burtoni, DCX is very useful for identifying new neurons in the transition zone of the retina. In addition, this marker can be applied to follow axons from maturing neurons through the neural fiber layer, optic nerve head, and optic nerve.

Cultures of glial cells from optic nerve of two adult teleost fish: Astatotilapia burtoni and Danio rerio

BACKGROUND

In vitro studies are very useful to increase the knowledge of different cell types and could be the key to understand cell metabolism and function. Fish optic nerves (ON) can recover visual functions by reestablishing its structure and reconnecting the axons of ganglion cells. This is because fish show spontaneous regeneration of the central nervous system which does not occur in mammals. In addition, several studies have indicated that glial cells of ON have different properties in comparison to the glial cells from brain or retina. Consequently, providing an in vitro tool will be highly beneficial to increase the knowledge of these cells.

NEW METHOD

We developed a cell culture protocol to isolate glial cells from ON of two teleost fish species, Danio rerio and Astatotilapia burtoni.

RESULTS

The optimized protocol allowed us to obtain ON cells and brain-derived cells from adult teleost fish. These cells were characterized as glial cells and their proprieties in vitro were analyzed.Comparison with Existing Method(s): Although it is striking that ON glial cells show peculiarities, their study in vitro has been limited by the only published protocol going back to the 1990s. Our protocol makes glial cells of different fish species available for experiments and studies to increase the understanding of these glial cell types.

CONCLUSIONS

This validated and effective in vitro tool increases the possibilities on studies of glial cells from fish ON which implies a reduction in animal experimentation.

Multiple roles for Pax2 in the embryonic mouse eye

The vertebrate eye anlage grows out of the brain and folds into bilayered optic cups. The eye is patterned along multiple axes, precisely controlled by genetic programs, to delineate neural retina, pigment epithelium, and optic stalk tissues. Pax genes encode developmental regulators of key morphogenetic events, with Pax2 being essential for interpreting inductive signals, including in the eye. PAX2 mutations cause ocular coloboma, when the ventral optic fissure fails to close. Previous studies established that Pax2 is necessary for fissure closure and to maintain the neural retina -- glial optic stalk boundary. Using a Pax2^GFP/+ knock-in allele we discovered that the mutant optic nerve head (ONH) lacks molecular boundaries with the retina and RPE, rendering the ONH larger than normal. This was preceded by ventronasal cup mispatterning, a burst of overproliferation and followed by optic cup apoptosis. Our findings support the hypothesis that ONH cells are tripotential, requiring Pax2 to remain committed to glial fates. This work extends current models of ocular development, contributes to broader understanding of tissue boundary formation and informs the underlying mechanisms of human coloboma.

GFAP expression in the optic nerve and increased H2S generation in the integration centers of the rainbow trout (Oncorhynchus mykiss) brain after unilateral eye injury

Hydrogen sulfide (H₂S) is considered as a protective factor against cardiovascular disorders. However, there are few reports on the effects of H₂S in the central nervous system during stress or injury. Previous studies on goldfish have shown that astrocytic response occurs in the damaged and contralateral optic nerves. Glial fibrillary acidic protein (GFAP) concentration in the optic nerves of rainbow trout has not been measured previously. This study further characterized the astrocytic response in the optic nerve and the brain of a rainbow trout (Oncorhynchus mykiss) after unilateral eye injury and estimated the amount of H₂S-producing enzyme cystathionine β-synthase (CBS) in the brain of the rainbow trout. Within 1 week after unilateral eye injury, a protein band corresponding to a molecular weight of 50 kDa was identified in the ipsi- and contralateral optic nerves of the rainbow trout. The concentration of GFAP in the injured optic nerve increased compared to the protein concentration on the contralateral side. The results of a quantitative analysis of GFAP⁺ cell distribution in the contralateral optic nerve showed the largest number of GFAP⁺ cells and fibers in the optic nerve head. In the damaged optic nerve, patterns of GFAP⁺ cell migration and large GFAP⁺ bipolar activated astrocytes were detected at 1 week after unilateral eye injury. The study of H₂S-producing system after unilateral eye injury in the rainbow trout was conducted using enzyme-linked immunosorbent assay, western blot analysis, and immunohistochemistry of polyclonal antibodies against CBS in the integrative centers of the brain: telencephalon, optic tectum, and cerebellum. Enzyme-linked immunosorbent assay results showed a 1.7-fold increase in CBS expression in the rainbow trout brain at 1 week after unilateral eye injury compared with that in intact animals. In the ventricular and subventricular regions of the rainbow trout telencephalon, CBS⁺ radial glia and neuroepithelial cells were identified. After unilateral eye injury, the number of CBS⁺ neuroepithelial cells in the pallial and subpallial periventricular regions of the telencephalon increased. In the optic tectum, unilateral eye injury led to an increase in CBS expression in radial glial cells; simultaneously, the number of CBS⁺ neuroepithelial cells decreased in intact animals. In the cerebellum of the rainbow trout, neuroglial interrelationships were revealed, where H₂S was released, apparently, from astrocyte-like cells. The organization of H₂S-producing cell complexes suggests that, the amount of glutamate produced in the rainbow trout cerebellum and its reuptake was controlled by astrocyte-like cells, reducing its excitotoxicity. In the dorsal matrix zone and granular eminences of the rainbow trout cerebellum, CBS was expressed in neuroepithelial cells. After unilateral eye injury, the level of CBS activity increased in all parts of the cerebellum. An increase in the number of H₂S-producing cells was a response to oxidative stress after unilateral eye injury, and the overproduction of H₂S in the cerebellum occurred to neutralize reactive oxygen species, providing the cells of the rainbow trout cerebellum with a protective effect. A structural reorganization in the dorsal matrix zone, associated with the appearance of an additional CBS⁺ apical zone, and a decrease in the enzyme activity in the dorsal matrix zone, was revealed in the zones of constitutive neurogenesis. All experiments were approved by the Commission on Biomedical Ethics, A.V. Zhirmunsky National Scientific Center of Marine Biology (NSCMB), Far Eastern Branch, Russian Academy of Science (FEB RAS) (approval No. 1) on July 31, 2019.

Expression of Developmentally Important Axon Guidance Cues in the Adult Optic Chiasm

Purpose

Regeneration of optic nerve axons after injury can be facilitated by several approaches, but misguidance at the optic chiasm is often observed. We characterized guidance cues in the embryonic visual system and adult optic chiasm before and after optic nerve crush (ONC) injury to better understand barriers to optic nerve regeneration in adults.

Methods

Radial glial (RC2/BLBP/Slit1), developmental (Pax2) and extracellular markers (CSPG: H2B/CS-56) were assessed in C57BL/6J mice by immunohistochemistry. RC2, BLBP, Slit1, and CSPG are known inhibitory guidance cues while Pax2 is a permissive guidance cue.

Results

At embryonic day 15.5 (E.15.5), RC2 and BLBP were identified superior to, and extending through, the optic chiasm. The optic chiasm was BLBP-ve in adult uninjured mice but BLBP+ve in adult mice 10 days after ONC injury. The reverse was true for RC2. Both BLBP and RC2 were absent in adult mice 6 weeks post-ONC. Slit1 was present in the optic chiasm midline and optic tracts in embryonic samples but was absent in uninjured adult tissue. Slit1 was observed superior to and at the midline of the optic chiasm 10 days post-ONC but absent 6 weeks after injury. Pax2 was expressed at the junction between the optic nerve and optic chiasm in embryonic brain tissue. In embryonic sections, CS-56 was observed at the junction between the optic chiasm and optic tract, and immediately superior to the optic chiasm. Both 2H6 and CS-56 staining was absent in uninjured and ONC-injured adult brains.

Conclusion

Differences in guidance cue expression during development, in adulthood and after injury may contribute to misguidance of regenerating RGC axons in the adult optic chiasm.

Neurolin expression in the optic nerve and immunoreactivity of Pax6-positive niches in the brain of rainbow trout (Oncorhynchus mykiss) after unilateral eye injury

In contrast to astrocytes in mammals, fish astrocytes promote axon regeneration after brain injury and actively participate in the regeneration process. Neurolin, a regeneration-associated, Zn8-labeled protein, is involved in the repair of damaged optic nerve in goldfish. At 1 week after unilateral eye injury, the expression of neurolin in the optic nerve and chiasm, and the expression of Pax6 that influences nervous system development in various brain regions in the rainbow trout (Oncorhynchus mykiss) were detected. Immunohistochemical staining revealed that the number of Zn8⁺ cells in the optic nerve head and intraorbital segment was obviously increased, and the increase in Zn8⁺ cells was also observed in the proximal and distal parts of injured optic nerve. This suggests that Zn8⁺ astrocytes participate in optic nerve regeneration. ELISA results revealed that Pax6 protein increased obviously at 1 week post-injury. Immunohistochemical staining revealed the appearance of Pax6⁺ neurogenic niches and a larger number of neural precursor cells, which are mainly from Pax6⁺ radial glia cells, in the nuclei of the diencephalon and optic tectum of rainbow trout (Oncorhynchus mykiss). Taken together, unilateral eye injury can cause optic nerve reaction, and the formation of neurogenic niches is likely a compensation phenomenon during the repair process of optic nerve injury in rainbow trout (Oncorhynchus mykiss).

Sox10 Expression in Goldfish Retina and Optic Nerve Head in Controls and after the Application of Two Different Lesion Paradigms

The mammalian central nervous system (CNS) is unable to regenerate. In contrast, the CNS of fish, including the visual system, is able to regenerate after damage. Moreover, the fish visual system grows continuously throughout the life of the animal, and it is therefore an excellent model to analyze processes of myelination and re-myelination after an injury. Here we analyze Sox10⁺ oligodendrocytes in the goldfish retina and optic nerve in controls and after two kinds of injuries: cryolesion of the peripheral growing zone and crushing of the optic nerve. We also analyze changes in a major component of myelin, myelin basic protein (MBP), as a marker for myelinated axons. Our results show that Sox10⁺ oligodendrocytes are located in the retinal nerve fiber layer and along the whole length of the optic nerve. MBP was found to occupy a similar location, although its loose appearance in the retina differed from the highly organized MBP⁺ axon bundles in the optic nerve. After optic nerve crushing, the number of Sox10⁺ cells decreased in the crushed area and in the optic nerve head. Consistent with this, myelination was highly reduced in both areas. In contrast, after cryolesion we did not find changes in the Sox10⁺ population, although we did detect some MBP^- degenerating areas. We show that these modifications in Sox10⁺ oligodendrocytes are consistent with their role in oligodendrocyte identity, maintenance and survival, and we propose the optic nerve head as an excellent area for research aimed at better understanding of de- and remyelination processes.

Cell proliferation and apoptosis in optic nerve and brain integration centers of adult trout Oncorhynchus mykiss after optic nerve injury

Fishes have remarkable ability to effectively rebuild the structure of nerve cells and nerve fibers after central nervous system injury. However, the underlying mechanism is poorly understood. In order to address this issue, we investigated the proliferation and apoptosis of cells in contralateral and ipsilateral optic nerves, after stab wound injury to the eye of an adult trout Oncorhynchus mykiss. Heterogenous population of proliferating cells was investigated at 1 week after injury. TUNEL labeling gave a qualitative and quantitative assessment of apoptosis in the cells of optic nerve of trout 2 days after injury. After optic nerve injury, apoptotic response was investigated, and mass patterns of cell migration were found. The maximal concentration of apoptotic bodies was detected in the areas of mass clumps of cells. It is probably indicative of massive cell death in the area of high phagocytic activity of macrophages/microglia. At 1 week after optic nerve injury, we observed nerve cell proliferation in the trout brain integration centers: the cerebellum and the optic tectum. In the optic tectum, proliferating cell nuclear antigen (PCNA)-immunopositive radial glia-like cells were identified. Proliferative activity of nerve cells was detected in the dorsal proliferative (matrix) area of the cerebellum and in parenchymal cells of the molecular and granular layers whereas local clusters of undifferentiated cells which formed neurogenic niches were observed in both the optic tectum and cerebellum after optic nerve injury. In vitro analysis of brain cells of trout showed that suspension cells compared with monolayer cells retain higher proliferative activity, as evidenced by PCNA immunolabeling. Phase contrast observation showed mitosis in individual cells and the formation of neurospheres which gradually increased during 1-4 days of culture. The present findings suggest that trout can be used as a novel model for studying neuronal regeneration.

Human Bone Marrow Stromal Cells Differentiate into Corneal Tissue and Prevent Ocular Graft-Versus-Host Disease in Mice

Clinical trials have assessed the use of human bone marrow stromal cells (hBMSCs) for the treatment of immune-related disorders such as graft-versus-host disease (GVHD). In the current study, we show that GFP+-transduced hBMSCs generated from bone marrow migrate and differentiate into corneal tissue after subconjunctival injection in mice. Interestingly, these hBMSCs display morphological features of epithelial, stromal, and endothelial cells and appear at different layers and with different morphologies depending on their position within the epithelium. Furthermore, these cells display ultrastructural properties, such as bundles of intermediate filaments, interdigitations, and desmosomes with GFP- cells, which confirms their differentiation into corneal tissues. GFP+-transduced hBMSCs were injected at different time points into the right eye of lethally irradiated mice undergoing bone marrow transplantation, which developed ocular GVHD (oGVHD). Remarkably, hBMSCs massively migrate to corneal tissues after subconjunctival injection. Both macroscopic and histopathological examination showed minimal or no evidence of GVHD in the right eye, while the left eye, where no hBMSCs were injected, displayed features of GVHD. Thus, in the current study, we confirm that hBMSCs may induce their therapeutic effect at least in part by differentiation and regeneration of damaged tissues in the host. Our results provide experimental evidence that hBMSCs represent a potential cellular therapy to attenuate oGVHD.

Regrowth of transected retinal ganglion cell axons despite persistent astrogliosis in the lizard (Gallotia galloti)

We analysed the astroglia response that is concurrent with spontaneous axonal regrowth after optic nerve (ON) transection in the lizard Gallotia galloti. At different post-lesional time points (0.5, 1, 3, 6, 9 and 12 months) we used conventional electron microscopy and specific markers for astrocytes [glial fibrillary acidic protein (GFAP), vimentin (Vim), sex-determining region Y-box-9 (Sox9), paired box-2 (Pax2)¸ cluster differentiation-44 (CD44)] and for proliferating cells (PCNA). The experimental retina showed a limited glial response since the increase of gliofilaments was not significant when compared with controls, and proliferating cells were undetectable. Conversely, PCNA(+) cells populated the regenerating ON, optic tract (OTr) and ventricular wall of both the hypothalamus and optic tectum (OT). Subpopulations of these PCNA(+) cells were identified as GFAP(+) and Vim(+) reactive astrocytes and radial glia. Reactive astrocytes up-regulated Vim at 1 month post-lesion, and both Vim and GFAP at 12 months post-lesion in the ON-OTr, indicating long-term astrogliosis. They also expressed Pax2, Sox9 and CD44 in the ON, and Sox9 in the OTr. Concomitantly, persistent tissue cavities and disorganised regrowing fibre bundles reaching the OT were observed. Our ultrastructural data confirm abundant gliofilaments in reactive astrocytes joined by desmosomes. Remarkably, they also accumulated myelin debris and lipid droplets until late stages, indicating their participation in myelin removal. These data suggest that persistent mammalian-like astrogliosis in the adult lizard ON contributes to a permissive structural scaffold for long-term axonal regeneration and provides a useful model to study the molecular mechanisms involved in these beneficial neuron-glia interactions.

Neuronal and glial differentiation during lizard (Gallotia galloti) visual system ontogeny

We studied the histogenesis of the lizard visual system (E30 to adulthood) by using a selection of immunohistochemical markers that had proved relevant for other vertebrates. By E30, the Pax6(+) pseudostratified retinal epithelium shows few newborn retinal ganglion cells (RGCs) in the centrodorsal region expressing neuron- and synaptic-specific markers such as betaIII-tubulin (Tuj1), synaptic vesicle protein-2 (SV2), and vesicular glutamate transporter-1 (VGLUT1). Concurrently, pioneer RGC axons run among the Pax2(+) astroglia in the optic nerve and reach the superficial optic tectum. Between E30 and E35, the optic chiasm and optic tract remain acellular, but the latter contains radial processes with subpial endfeet expressing vimentin (Vim). From E35, neuron- and synaptic-specific stainings spread in the retina and optic tectum, whereas retinal Pax6, and Tuj1/SV2 in RGC axons decrease. Müller glia and abundant optic nerve glia express a variety of glia-specific markers until adulthood. Subpopulations of optic nerve glia are also VGLUT1(+) and cluster differentiation-44 (CD44)-positive but cytokeratin-negative, unlike the case in other regeneration-competent species. Specifically, coexpression of CD44/Vim and glutamine synthetase (GS)/VGLUT1 reflects glial specialization, insofar as most CD44(+) glia are GS(-). In the adult optic tract and tectum, radial glia and free astroglia coexist. The latter show different immunocharacterization (Pax2(-)/CD44(-) /Vim(-)) compared with that in the optic nerve. We conclude that upregulation of Tuj1 and SV2 is required for axonal outgrowth and search for appropriate targets, whereas Pax2(+) optic nerve astroglia and Vim(+) radial glia may aid in early axonal guidance. Spontaneous axonal regrowth seems to succeed despite the heterogeneous mammalian-like glial environment in the lizard optic nerve.

Characterization of Pax2 expression in the goldfish optic nerve head during retina regeneration

The Pax2 transcription factor plays a crucial role in axon-guidance and astrocyte differentiation in the optic nerve head (ONH) during vertebrate visual system development. However, little is known about its function during regeneration. The fish visual system is in continuous growth and can regenerate. Müller cells and astrocytes of the retina and ONH play an important role in these processes. We demonstrate that pax2a in goldfish is highly conserved and at least two pax2a transcripts are expressed in the optic nerve. Moreover, we show two different astrocyte populations in goldfish: Pax2⁺ astrocytes located in the ONH and S100⁺ astrocytes distributed throughout the retina and the ONH. After peripheral growth zone (PGZ) cryolesion, both Pax2⁺ and S100⁺ astrocytes have different responses. At 7 days after injury the number of Pax2⁺ cells is reduced and coincides with the absence of young axons. In contrast, there is an increase of S100⁺ astrocytes in the retina surrounding the ONH and S100⁺ processes in the ONH. At 15 days post injury, the PGZ starts to regenerate and the number of S100⁺ astrocytes increases in this region. Moreover, the regenerating axons reach the ONH and the pax2a gene expression levels and the number of Pax2⁺ cells increase. At the same time, S100⁺/GFAP⁺/GS⁺ astrocytes located in the posterior ONH react strongly. In the course of the regeneration, Müller cell vitreal processes surrounding the ONH are primarily disorganized and later increase in number. During the whole regenerative process we detect a source of Pax2⁺/PCNA⁺ astrocytes surrounding the posterior ONH. We demonstrate that pax2a expression and the Pax2⁺ astrocyte population in the ONH are modified during the PGZ regeneration, suggesting that they could play an important role in this process.

Astroglial structures in the zebrafish brain

To understand components shaping the neuronal environment we studied the astroglial cells in the zebrafish brain using immunocytochemistry for structural and junctional markers, electron microscopy including freeze fracturing, and probed for the water channel protein aquaporin-4. Glial fibrillary acidic protein (GFAP) and glutamine synthetase (GS) showed largely overlapping immunoreactivity: GFAP in the main glial processes and GS in main processes and smaller branches. Claudin-3 immunoreactivity was spread in astroglial cells along their major processes. The ventricular lining was immunoreactive for the tight-junction associated protein ZO-1, in the telencephalon located on the dorsal, lateral, and medial surface due to the everting morphogenesis. In the tectum, subpial glial endfeet were also positive for ZO-1. Correspondingly, electron microscopy revealed junctional complexes between subpial glial endfeet. However, in freeze-fracture analysis tight junctional strands were not found between astroglial membranes, either in the optic tectum or in the telencephalon. Occurrence of aquaporin-4, the major astrocytic water channel in mammals, was demonstrated by polymerase chain reaction (PCR) analysis and immunocytochemistry in tectum and telencephalon. Localization of aquaporin-4 was not polarized but distributed along the entire radial extent of the cell. Interestingly, their membranes were devoid of the orthogonal arrays of particles formed by aquaporin-4 in mammals. Finally, we investigated astroglial cells in proliferative areas. Brain lipid basic protein, a marker of early glial differentiation but not GS, were present in some proliferation zones, whereas cells lining the ventricle were positive for both markers. Thus, astroglial cells in the zebrafish differ in many aspects from mammalian astrocytes.

Pax2 is expressed in a subpopulation of Müller cells in the central chick retina

Müller cells in the chick retina are generally thought to be a homogeneous population. We show that the transcription factor Pax2 is expressed by Müller cells in the central chick retina and its expression was first observed at stage 32 (embryonic day [E] 7.5). Birth-dating indicated that the majority of Pax2-positive Müller cells are generated between stage 29 and 33 (E5.5-E8). At stage 42 (E16), several Müller cell markers, such as Sox2 and 2M6, had reached the peripheral retina, while the Pax2 labeling extended approximately half-way. A similar pattern was maintained in the 6-month-old chicken. Neither the Pax2-positive nor the Pax2-negative Müller cells could be specifically associated to proliferative responses in the retina induced by growth factors or N-methyl-D-aspartate. Pax2 was not detected in Müller cells in mouse, rat, guinea-pig, rabbit, or pig retinas; but the zebrafish retina displayed a similar pattern of central Pax2-expressing Müller cells.

Comparative study of Pax2 expression in glial cells in the retina and optic nerve of birds and mammals

Little is known about the expression of Pax2 in mature retina or optic nerve. Here we probed for the expression of Pax2 in late stages of embryonic development and in mature chick retina. We find two distinct Pax2 isoforms expressed by cells within the retina and optic nerve. Surprisingly, Müller glia in central regions of the retina express Pax2, and levels of expression are decreased with increasing distance from the nerve head. In Müller glia, the expression levels of Pax2 are increased by acute retinal damage or treatment with growth factors. At the optic nerve, Pax2 is expressed by peripapillary glia, at the junction of the neural retina and optic nerve head and by glia within the optic nerve. In addition, we assayed for Pax2 expression in glial cells in mammalian retinas. In mammalian retinas, unlike the case in chick retina, the Müller glia do not express Pax2. Pax2-expressing cells are found in the optic nerve and astrocytes within the mouse retina. By comparison, Pax2-positive cells are not found within the guinea pig retina; Pax2-expressing glia are confined to the optic nerve. In dog and monkey (Macaca fascicularis), Pax2 is expressed by astrocytes that are scattered across inner retinal layers and by numerous glia within the optic nerve. Interestingly, Pax2-positive glial cells are found at the peripheral edge of the dog retina, but only in older animals. We conclude that the expression of Pax2 in the vertebrate eye is restricted to retinal astrocytes, peripapillary glia, and glia within the optic nerve.