Axotomy-induced retinal ganglion cell death in adult mice: quantitative and topographic time course analyses

REDD2-mediated inhibition of mTOR promotes dendrite retraction induced by axonal injury

Dendritic defects occur in neurodegenerative diseases accompanied by axonopathy, yet the mechanisms that regulate these pathologic changes are poorly understood. Using Thy1-YFPH mice subjected to optic nerve axotomy, we demonstrate early retraction of retinal ganglion cell (RGC) dendrites and selective loss of mammalian target of rapamycin (mTOR) activity, which precede soma loss. Axonal injury triggered rapid upregulation of the stress-induced protein REDD2 (regulated in development and DNA damage response 2), a potent inhibitor of mTOR. Short interfering RNA-mediated REDD2 knockdown restored mTOR activity and rescued dendritic length, area and branch complexity in a rapamycin-dependent manner. Whole-cell recordings demonstrated that REDD2 depletion leading to mTOR activation in RGCs restored their light response properties. Lastly, we show that REDD2-dependent mTOR activity extended RGC survival following axonal damage. These results indicate that injury-induced stress leads to REDD2 upregulation, mTOR inhibition and dendrite pathology causing neuronal dysfunction and subsequent cell death.

Nerve fiber layer thinning lags retinal ganglion cell density following crush axonopathy

PURPOSE

We investigated the progressive nature of neurodegenerative structural changes following injury to retinal ganglion cell (RGC) axons using quantifiable and noninvasive in vivo imaging techniques.

METHODS

To track degenerative RGC progression in retinas following optic nerve crush (ONC) injury, spectral-domain optical coherence tomography (SD-OCT) was used to quantitate the RGC nerve fiber layer (NFL) density. The RGC soma cell density (RCD) was measured by confocal scanning laser ophthalmoscopy (CSLO). The RCD counts were performed using blood vessels as landmarks to anatomically track defined progressive changes in enhanced yellow fluorescent fusion protein (EYFP)-labeled RGCs.

RESULTS

Following ONC injury, 68% of the observed decrease in RCD measured by CSLO and 54% of the NFL thickness obtained by SD-OCT imaging (N=4 retinas) occurred within the first week. Between days 7 and 14, an additional 22% decrease in RCD was concurrent with a 31% decrease in overall NFL thickness. Finally, between days 14 and 21, an additional 10% decrease in RCD measured in vivo by CSLO and 15% decrease in NFL thickness by SD-OCT was observed.

CONCLUSIONS

Our data suggest that in vivo CSLO imaging of EYFP-RGC expression and SD-OCT measured NFL thickness are fast and reliable methods that longitudinally track neurodegenerative progression following ONC injury. Neurodegenerative changes in NFL thickness measured by SD-OCT imaging have the same overall trajectory as those observed by CSLO for RCD; however, changes in NFL thickness initially lag behind in vivo RGC soma counts with a slower decline in overall measurable change.

Retinal vessel diameter changes induced by transient high perfusion pressure

AIM

To investigate the effects of transient high perfusion pressure on the retinal vessel diameter and retinal ganglion cells.

METHODS

The animals were divided into four groups according to different infusion pressure and infusion time (60 mm Hg-3min, 60 mm Hg-5min, 100 mm Hg-3min, 100 mm Hg-5min). Each group consisted of six rabbits. The left eye was used as the experimental eye and the right as a control. Retinal vascular diameters were evaluated before, during infusion, immediately after infusion, 5min, 10min and 30min after infusion based on the fundus photographs. Blood pressure was monitored during infusion. The eyes were removed after 24h. Damage to retinal ganglion cell (RGC) was analyzed by histology.

RESULTS

Retina became whiten and papilla optic was pale during perfusion. Measurements showed significant decrease in retinal artery and vein diameter during perfusion in all of the four groups at the proximal of the edge of the optic disc. The changes were significant in the 100 mm Hg-3min group and 100 mm Hg-5min group compared with 60 mm Hg-3min group (P1=0.025, P2=0.000). The diameters in all the groups recovered completely after 30min of reperfusion. The number of RGC showed no significant changes at the IOP in 100 mm Hg with 5min compared with contralateral untreated eye (P>0.05).

CONCLUSION

Transient fluctuations during infusion lead to temporal changes of retinal vessels, which could affect the retinal blood circulation. The RGCs were not affected by this transient fluctuation. Further studies are necessary to evaluate the effect of pressure during real-time phacoemusification on retinal blood circulation.

Number and distribution of mouse retinal cone photoreceptors: differences between an albino (Swiss) and a pigmented (C57/BL6) strain

We purpose here to analyze and compare the population and topography of cone photoreceptors in two mouse strains using automated routines, and to design a method of retinal sampling for their accurate manual quantification. In whole-mounted retinas from pigmented C57/BL6 and albino Swiss mice, the longwave-sensitive (L) and the shortwave-sensitive (S) opsins were immunodetected to analyze the population of each cone type. In another group of retinas both opsins were detected with the same fluorophore to quantify all cones. In a third set of retinas, L-opsin and Brn3a were immunodetected to determine whether L-opsin⁺cones and retinal ganglion cells (RGCs) have a parallel distribution. Cones and RGCs were automatically quantified and their topography illustrated with isodensity maps. Our results show that pigmented mice have a significantly higher number of total cones (all-cones) and of L-opsin⁺cones than albinos which, in turn, have a higher population of S-opsin⁺cones. In pigmented animals 40% of cones are dual (cones that express both opsins), 34% genuine-L (cones that only express the L-opsin), and 26% genuine-S (cones that only express the S-opsin). In albinos, 23% of cones are genuine-S and the proportion of dual cones increases to 76% at the expense of genuine-L cones. In both strains, L-opsin⁺cones are denser in the central than peripheral retina, and all-cones density increases dorso-ventrally. In pigmented animals S-opsin⁺cones are scarce in the dorsal retina and very numerous in the ventral retina, being densest in its nasal aspect. In albinos, S-opsin⁺cones are abundant in the dorsal retina, although their highest densities are also ventral. Based on the densities of each cone population, we propose a sampling method to manually quantify and infer their total population. In conclusion, these data provide the basis to study cone degeneration and its prevention in pathologic conditions.

Optic nerve crush induces spatial and temporal gene expression patterns in retina and optic nerve of BALB/cJ mice

Background

Central nervous system (CNS) trauma and neurodegenerative disorders trigger a cascade of cellular and molecular events resulting in neuronal apoptosis and regenerative failure. The pathogenic mechanisms and gene expression changes associated with these detrimental events can be effectively studied using a rodent optic nerve crush (ONC) model. The purpose of this study was to use a mouse ONC model to: (a) evaluate changes in retina and optic nerve (ON) gene expression, (b) identify neurodegenerative pathogenic pathways and (c) discover potential new therapeutic targets.

Results

Only 54% of total neurons survived in the ganglion cell layer (GCL) 28 days post crush. Using Bayesian Estimation of Temporal Regulation (BETR) gene expression analysis, we identified significantly altered expression of 1,723 and 2,110 genes in the retina and ON, respectively. Meta-analysis of altered gene expression (≥1.5, ≤-1.5, p < 0.05) using Partek and DAVID demonstrated 28 up and 20 down-regulated retinal gene clusters and 57 up and 41 down-regulated optic nerve clusters. Regulated gene clusters included regenerative change, synaptic plasticity, axonogenesis, neuron projection, and neuron differentiation. Expression of selected genes (Vsnl1, Syt1, Synpr and Nrn1) from retinal and ON neuronal clusters were quantitatively and qualitatively examined for their relation to axonal neurodegeneration by immunohistochemistry and qRT-PCR.

Conclusion

A number of detrimental gene expression changes occur that contribute to trauma-induced neurodegeneration after injury to ON axons. Nrn1 (synaptic plasticity gene), Synpr and Syt1 (synaptic vesicle fusion genes), and Vsnl1 (neuron differentiation associated gene) were a few of the potentially unique genes identified that were down-regulated spatially and temporally in our rodent ONC model. Bioinformatic meta-analysis identified significant tissue-specific and time-dependent gene clusters associated with regenerative changes, synaptic plasticity, axonogenesis, neuron projection, and neuron differentiation. These ONC induced neuronal loss and regenerative failure associated clusters can be extrapolated to changes occurring in other forms of CNS trauma or in clinical neurodegenerative pathological settings. In conclusion, this study identified potential therapeutic targets to address two key mechanisms of CNS trauma and neurodegeneration: neuronal loss and regenerative failure.

Assessment of inner retina dysfunction and progressive ganglion cell loss in a mouse model of glaucoma

The DBA/2J mouse is a model of ocular hypertension and retinal ganglion cell (RGC) degeneration, the main features of which are iris pigment dispersion (IPD) and iris stromal atrophy (ISA). These animals also experience glaucomatous changes, including an increase in intraocular pressure (IOP) beginning at about 9-12 months of age and sectorial RGC death in the retina. The aim of this study was to determine the onset of functional changes exhibited by DBA/2J mice in the inner retina. This was performed by means of electroretinographic recordings (scotopic threshold response, STR) and their correlation with morphological changes (loss of RGCs). To this end, we recorded the scotopic threshold response in control C57BL/6J and in DBA/2J mice at different ages. The RGCs, in both DBA/2J and C57BL/6J animals, were identified at 15 months of age by retrograde tracing with an analogue of fluorogold, hydroxystilbamidine methanesulfonate (OHSt), applied on the superior colliculi. Whole mount retinas were processed to quantify the population of RGCs identified by fluorogold tracing and Brn3a immunodetection, and were counted using image analysis software; an isodensity contour plot was generated for each retina. DBA/2J mice showed a significant reduction in the positive STR (pSTR) amplitudes at 12 months of age, as compared to control C57BL/6J mice of the same age. The pSTR mean amplitude decreased to approximately 27.82% of the values recorded in control mice (p = 0.0058). STR responses decreased in both strains as a result of the natural process of aging, but the decrease was more pronounced in DBA/2J mice. Furthermore, quantification of the total number of RGCs identified by OHSt and Brn3a expression showed a reduced population of RGCs in DBA/2J mice as compared to control mice. Regression analysis revealed significant correlations between the decrease in pSTR and a non-homogeneous reduction in the number of RGCs throughout the retina. Our results indicate the existence of a correlation between retinal function impairment and RGC loss. This functional and morphological analysis allows a reliable assessment of the progression of the disease.

Diffuse traumatic axonal injury in the optic nerve does not elicit retinal ganglion cell loss

Much of the morbidity after traumatic brain injury (TBI) is associated with traumatic axonal injury (TAI). Although most TAI studies focus on corpus callosum white matter, the visual system has received increased interest. To assess visual system TAI, we developed a mouse model of optic nerve TAI. It is unknown, however, whether this TAI causes retinal ganglion cell (RGC) death. To address this issue, YFP (yellow fluorescent protein)-16 transgenic mice were subjected to mild TBI and followed from 2 to 28 days. Neither TUNEL (terminal deoxynucleotidyl transferase dUTP nick end labeling)-positive or cleaved caspase-3-immunoreactive RGCs were observed from 2 to 28 days after TBI. Quantification of immunoreactivity of Brn3a, an RGC marker, demonstrated no RGC loss; parallel electron microscopic analysis confirmed RGC viability. Persistent RGC survival was also consistent with the finding of reorganization in the proximal axonal segments after TAI, wherein microglia/macrophages remained inactive. In contrast, activated microglia/macrophages closely enveloped the distal disconnected, degenerating axonal segments at 7 to 28 days after injury, thereby confirming that this model consistently evoked TAI followed by disconnection. Collectively, these data provide novel insight into the evolving pathobiology associated with TAI that will form a foundation for future studies exploring TAI therapy and its downstream consequences.

Number and spatial distribution of intrinsically photosensitive retinal ganglion cells in the adult albino rat

Intrinsically photosensitive retinal ganglion cells (ipRGCs) respond directly to light and are responsible of the synchronization of the circadian rhythm with the photic stimulus and for the pupillary light reflex. To quantify the total population of rat-ipRGCs and to assess their spatial distribution we have developed an automated routine and used neighbour maps. Moreover, in all analysed retinas we have studied the general population of RGCs - identified by their Brn3a expression - and the population of ipRGCs - identified by melanopsin immunodetection - thus allowing the co-analysis of their topography. Our results show that the total mean number ± standard deviation of ipRGCs in the albino rat is 2047 ± 309. Their distribution in the retina seems to be complementary to that of Brn3a(+)RGCs, being denser in the periphery, especially in the superior retina where their highest densities are found in the temporal quadrant, above the visual streak. In addition, by tracing the retinas from both superior colliculi, we have also determined that 90.62% of the ipRGC project to these central targets.

Whole number, distribution and co-expression of brn3 transcription factors in retinal ganglion cells of adult albino and pigmented rats

The three members of the Pou4f family of transcription factors: Pou4f1, Pou4f2, Pou4f3 (Brn3a, Brn3b and Brn3c, respectively) play, during development, essential roles in the differentiation and survival of sensory neurons. The purpose of this work is to study the expression of the three Brn3 factors in the albino and pigmented adult rat. Animals were divided into these groups: i) untouched; ii) fluorogold (FG) tracing from both superior colliculli; iii) FG-tracing from one superior colliculus; iv) intraorbital optic nerve transection or crush. All retinas were dissected as flat-mounts and subjected to single, double or triple immunohistofluorescence The total number of FG-traced, Brn3a, Brn3b, Brn3c or Brn3 expressing RGCs was automatically quantified and their spatial distribution assessed using specific routines. Brn3 factors were studied in the general RGC population, and in the intrinsically photosensitive (ip-RGCs) and ipsilateral RGC sub-populations. Our results show that: i) 70% of RGCs co- express two or three Brn3s and the remaining 30% express only Brn3a (26%) or Brn3b; ii) the most abundant Brn3 member is Brn3a followed by Brn3b and finally Brn3c; iii) Brn3 a-, b- or c- expressing RGCs are similarly distributed in the retina; iv) The vast majority of ip-RGCs do not express Brn3; v) The main difference between both rat strains was found in the population of ipsilateral-RGCs, which accounts for 4.2% and 2.5% of the total RGC population in the pigmented and albino strain, respectively. However, more ipsilateral-RGCs express Brn3 factors in the albino than in the pigmented rat; vi) RGCs that express only Brn3b and RGCs that co-express the three Brn3 members have the biggest nuclei; vii) After axonal injury the level of Brn3a expression in the surviving RGCs decreases compared to control retinas. Finally, this work strengthens the validity of Brn3a as a marker to identify and quantify rat RGCs.

Dopaminergic amacrine cells express opioid receptors in the mouse retina

The presence of opioid receptors has been confirmed by a variety of techniques in vertebrate retinas including those of mammals; however, in most reports, the location of these receptors has been limited to retinal regions rather than specific cell types. Concurrently, our knowledge of the physiological functions of opioid signaling in the retina is based on only a handful of studies. To date, the best-documented opioid effect is the modulation of retinal dopamine release, which has been shown in a variety of vertebrate species. Nonetheless, it is not known if opioids can affect dopaminergic amacrine cells (DACs) directly, via opioid receptors expressed by DACs. This study, using immunohistochemical methods, sought to determine whether (1) μ- and δ-opioid receptors (MORs and DORs, respectively) are present in the mouse retina, and if present, (2) are they expressed by DACs. We found that MOR and DOR immunolabeling were associated with multiple cell types in the inner retina, suggesting that opioids might influence visual information processing at multiple sites within the mammalian retinal circuitry. Specifically, colabeling studies with the DAC molecular marker anti-tyrosine hydroxylase antibody showed that both MOR and DOR immunolabeling localize to DACs. These findings predict that opioids can affect DACs in the mouse retina directly, via MOR and DOR signaling, and might modulate dopamine release as reported in other mammalian and nonmammalian retinas.

The Bcl-2 family member BIM has multiple glaucoma-relevant functions in DBA/2J mice

Axonal insult induces retinal ganglion cell (RGC) death through a BAX-dependent process. The pro-apoptotic Bcl-2 family member BIM is known to induce BAX activation. BIM expression increased in RGCs after axonal injury and its induction was dependent on JUN. Partial and complete Bim deficiency delayed RGC death after mechanical optic nerve injury. However, in a mouse model of glaucoma, DBA/2J mice, Bim deficiency did not prevent RGC death in eyes with severe optic nerve degeneration. In a subset of DBA/2J mice, Bim deficiency altered disease progression resulting in less severe nerve damage. Bim deficient mice exhibited altered optic nerve head morphology and significantly lessened intraocular pressure elevation. Thus, a decrease in axonal degeneration in Bim deficient DBA/2J mice may not be caused by a direct role of Bim in RGCs. These data suggest that BIM has multiple roles in glaucoma pathophysiology, potentially affecting susceptibility to glaucoma through several mechanisms.

Longitudinal in vivo imaging of retinal ganglion cells and retinal thickness changes following optic nerve injury in mice

BACKGROUND

Retinal ganglion cells (RGCs) die in sight-threatening eye diseases. Imaging RGCs in humans is not currently possible and proof of principle in experimental models is fundamental for future development. Our objective was to quantify RGC density and retinal thickness following optic nerve transection in transgenic mice expressing cyan fluorescent protein (CFP) under control of the Thy1 promoter, expressed by RGCs and other neurons.

METHODOLOGY/PRINCIPAL FINDINGS

A modified confocal scanning laser ophthalmoscopy (CSLO)/spectral-domain optical coherence tomography (SD-OCT) camera was used to image and quantify CFP+ cells in mice from the B6.Cg-Tg(Thy1-CFP)23Jrs/J line. SD-OCT circle (1 B-scan), raster (37 B-scans) and radial (24 B-scans) scans of the retina were also obtained. CSLO was performed at baseline (n = 11) and 3 (n = 11), 5 (n = 4), 7 (n = 10), 10 (n = 6), 14 (n = 7) and 21 (n = 5) days post-transection, while SD-OCT was performed at baseline and 7, 14 and 35 days (n = 9) post-transection. Longitudinal change in CFP+ cell density and retinal thickness were computed. Compared to baseline, the mean (SD) percentage CFP+ cells remaining at 3, 5, 7, 10, 14 and 21 days post-transection was 86 (9)%, 63 (11)%, 45 (11)%, 31 (9)%, 20 (9)% and 8 (4)%, respectively. Compared to baseline, the mean (SD) retinal thickness at 7 days post-transection was 97 (3)%, 98 (2)% and 97 (4)% for the circle, raster and radial scans, respectively. The corresponding figures at 14 and 35 days post-transection were 96 (3)%, 97 (2)% and 95 (3)%; and 93 (3)%, 94 (3)% and 92 (3)%.

CONCLUSIONS/SIGNIFICANCE

Longitudinal imaging showed an exponential decline in CFP+ cell density and a small (≤8%) reduction in SD-OCT measured retinal thickness post-transection. SD-OCT is a promising tool for detecting structural changes in experimental optic neuropathy. These results represent an important step towards translation for clinical use.

Traumatology of the optic nerve and contribution of crystallins to axonal regeneration

Within a few decades, the repair of long neuronal pathways such as spinal cord tracts, the optic nerve or intracerebral tracts has gone from being strongly contested to being recognized as a potential clinical challenge. Cut axonal stumps within the optic nerve were originally thought to retract and become irreversibly necrotic within the injury zone. Optic nerve astrocytes were assumed to form a gliotic scar and remodelling of the extracellular matrix to result in a forbidden environment for re-growth of axons. Retrograde signals to the ganglion cell bodies were considered to prevent anabolism, thus also initiating apoptotic death and gliotic repair within the retina. However, increasing evidence suggests the reversibility of these regressive processes, as shown by the analysis of molecular events at the site of injury and within ganglion cells. We review optic nerve repair from the perspective of the proximal axon stump being a major player in determining the successful formation of a growth cone. The axonal stump and consequently the prospective growth cone, communicates with astrocytes, microglial cells and the extracellular matrix via a panoply of molecular tools. We initially highlight these aspects on the basis of recent data from numerous laboratories. Then, we examine the mechanisms by which an injury-induced growth cone can sense its surroundings within the area distal to the injury. Based on requirements for successful axonal elongation within the optic nerve, we explore the models employed to instigate successful growth cone formation by ganglion cell stimulation and optic nerve remodelling, which in turn accelerate growth. Ultimately, with regard to the proteomics of regenerating retinal tissue, we discuss the discovery of isoforms of crystallins, with crystallin beta-b2 (crybb2) being clearly upregulated in the regenerating retina. Crystallins are produced and used to promote the elongation of growth cones. In vivo and in vitro, crystallins beta and gamma additionally promote the growth of axons by enhancing the production of ciliary neurotrophic factor (CNTF), indicating that they also act on astrocytes to promote axonal regrowth synergistically. These are the first data showing that axonal regeneration is related to crybb2 movement within neurons and to additional stimulation of CNTF. We demonstrate that neuronal crystallins constitute a novel class of neurite-promoting factors that probably operate through an autocrine and paracrine mechanism and that they can be used in neurodegenerative diseases. Thus, the post-injury fate of neurons cannot be seen merely as inevitable but, instead, must be regarded as a challenge to shape conditions for initiating growth cone formation to repair the damaged optic nerve.