David Hicks

Rat photoreceptor circadian oscillator strongly relies on lighting conditions.

Mammalian retina harbours a self‐sustained circadian clock able to synchronize to the lightu2003:u2003dark (LD) cycle and to drive cyclic outputs such as night‐time melatonin synthesis. Clock genes are expressed in distinct parts of the tissue, and it is presently assumed that the retina contains several circadian oscillators. However, molecular organization of cell type‐specific clockworks has been poorly investigated. Here, we questioned the presence of a circadian clock in rat photoreceptors by studying 24‐h kinetics of clock and clock output gene expression in whole photoreceptor layers isolated by vibratome sectioning. To address the importance of light stimulation towards photoreceptor clock properties, animals were exposed to 12u2003:u200312u2003h LD cycle or 36u2003h constant darkness. Clock, Bmal1, Per1, Per2, Cry1, Cry2, RevErbα and Rorβ clock genes were all found to be expressed in photoreceptors and to display rhythmic transcription in LD cycle. Clock genes in whole retinas, used as a reference, also showed rhythmic expression with marked similarity to the profiles in pure photoreceptors. In contrast, clock gene oscillations were no longer detectable in photoreceptor layers after 36u2003h darkness, with the exception of Cry2 and Rorβ. Importantly, transcripts from two well‐characterized clock output genes, Aanat (arylalkylamine N‐acetyltransferase) and c‐fos, retained sustained rhythmicity. We conclude that rat photoreceptors contain the core machinery of a circadian oscillator likely to be operative and to drive rhythmic outputs under exposure to a 24‐h LD cycle. Constant darkness dramatically alters the photoreceptor clockwork and circadian functions might then rely on inputs from extra‐photoreceptor oscillators.

Heterogeneity of intrinsically photosensitive retinal ganglion cells in the mouse revealed by molecular phenotyping

Intrinsically photosensitive retinal ganglion cell (ipRGC) types can be distinguished by their dendritic tree stratification and intensity of melanopsin staining. We identified heavily stained melanopsin‐positive M1 cells branching in the outermost part of the inner plexiform layer (IPL) and weakly melanopsin‐positive M2 cells branching in the innermost layer of the IPL. A third type can be distinguished by the displacement of the soma to the inner nuclear layer and has morphological similarities with either M1 cells or M2 cells, and is termed here displaced or M‐d cells. The aim of the present study was to examine the phenotypic traits of ipRGC types. Using whole retinae from adult mice, we performed immunohistochemistry using melanopsin immunostaining and a number of antibodies directed against proteins typically expressed in retinal ganglion cells. The majority of M1 and M2 ipRGCs expressed Isl‐1, microtubule associated protein‐2 (MAP2), γ‐synuclein, and NeuN, whereas Brn3 transcription factor and the different neurofilaments (NF68, NF160, NF200) were able to discriminate between ipRGC subtypes. Brn3 was expressed preferentially in M2 cells and in a small subpopulation of weakly melanopsin‐positive M‐d cells with similarities to M2 cells. All three neurofilaments were primarily expressed in large M2 cells with similarities to the recently described alpha‐like M4 cells, but not in M1 cells. Expression of NF68 and NF160 was also observed in a few large M‐d ipRGCs. These findings show that ipRGCs are not a phenotypically homogenous population and that specific neuronal markers (Brn3 and neurofilament) can partly distinguish between different ipRGC subtypes. J. Comp. Neurol. 521:912–932, 2013.

Diurnal rodents as animal models of human central vision: characterisation of the retina of the sand rat Psammomys obsesus

BackgroundCone photoreceptor-based central vision is of paramount importance in human eyesight, and the increasing numbers of persons affected by macular degeneration emphasizes the need for relevant and amenable animal models. Although laboratory mice and rats have provided valuable information on retinal diseases, they have inherent limitations for studies on macular pathology. In the present study, we extend our recent analyses of diurnal murid rodents to demonstrate that the sand rat Psammomys obesus has a remarkably cone-rich retina, and represents a useful adjunct to available animal models of central vision.MethodsAdult P. obesus were captured and transferred to animal facilities where they were maintained under standard light/dark cycles. Animals were euthanised and their eyes enucleated. Tissue was either fixed in paraformaldehyde and prepared for immunohistochemistry, or solubilized in lysis buffer and separated by SDS-PAGE and subjected to western blot analysis. Samples were labelled with a battery of antibodies against rod and cone photoreceptors, inner retinal neurones, and glia.ResultsP. obesus showed a high percentage of cones, 41% of total photoreceptor numbers in both central and peripheral retina. They expressed multiple cone-specific proteins, including short and medium-wavelength opsin and cone transducin. A second remarkable feature of the retina concerned the horizontal cells, which expressed high levels of glial fibrillar acidic protein and occludin, two proteins which are not seen in other species.ConclusionThe retina of P. obesus displays high numbers of morphologically and immunologically identifiable cones which will facilitate analysis of cone pathophysiology in this species. The unusual horizontal cell phenotype may be related to the cone distribution or to an alternative facet of the animals lifestyle.

Loss of photic entrainment at low illuminances in rats with acute photoreceptor degeneration.

In several species, an acute injection of N‐methyl‐N‐nitrosourea (MNU) induces a retinal degeneration characterized principally by a rapid loss of the outer nuclear layer, the other layers remaining structurally intact. It has, however, also been reported that down‐regulation of melanopsin gene expression is associated with the degeneration and is detectable soon after injection. Melanopsin is expressed by a small subset of intrinsically photosensitive retinal ganglion cells and plays an important role in circadian behaviour photoentrainment. We injected MNU into Long Evans rats and investigated the ability of animals to entrain to three light/dark cycles of different light intensities (300, 15 and 1 lux). Control animals entrained their locomotor activity rhythms to the three cycles. In contrast, MNU‐treated animals could only entrain properly to the 300 lux cycle. For the 15 lux cycle, their phase angle was much altered compared with control animals, and for the 1 lux cycle, MNU‐injected animals were unable to photoentrain and exhibited an apparent free‐run activity pattern with a period of 24.3u2003h. Subsequent to behavioural studies the animals were killed and rod, cone, melanopsin expression and melanopsin‐expressing cells were quantified. Rod and cone loss was almost complete, melanopsin protein was reduced by 83% and melanopsin‐expressing cells were reduced by 37%. Our study provides a comprehensive model of photoreceptor degeneration at the adult stage and a simple and versatile method to investigate the relation between retinal photoreceptors and the circadian system.

Retinas of the diurnal rodent Arvicanthis ansorgei are highly resistant to experimentally induced stress and degeneration.

PURPOSEnEnvironmentally induced stress plays a significant role in retinal degeneration and blindness both in animals and in humans. Among such sources of stress, phototoxicity is well studied and has been shown to lead to photoreceptor-specific loss in a number of species. However, the vast majority of studies have been conducted in nocturnal, albino rod-dominant rat and mouse strains, and the pertinence of such findings to human pathology and cone loss is debatable. The authors examined retinal vulnerability to damage in the diurnal murid rodent Arvicanthis ansorgei, a pigmented species with a large number of cones.nnnMETHODSnThe authors used established protocols for exposing animals to a wide range of lighting conditions (variable intensity, duration, spectrum, previous light history, and time of exposure) and injecting N-methyl-N-nitrosourea (MNU); each procedure is reported to produce rapid and complete photoreceptor-specific damage. Animals then underwent electroretinography to record rod and cone function and were subsequently euthanized and used for immunohistochemical analysis of retinal structure and quantification of free fatty acids.nnnRESULTSnThese standard regimens produced no detectable detrimental effects on A. ansorgei retinal phenotype, function, or structure. Partial retinal damage in A. ansorgei was induced by very intense blue light or elevated doses of MNU. This resistance was not attributable to differences in lipid composition (specifically, docosahexaenoic acid) between A. ansorgei and susceptible strains of mice and rats.nnnCONCLUSIONSnThe retina of this species exhibits exceptionally high resistance to damage from light and toxins such as MNU.

Intrinsic photosensitive retinal ganglion cells in the diurnal rodent, Arvicanthis ansorgei.

Intrinsically photosensitive retinal ganglion cells (ipRGCs) represent a new class of photoreceptors which support a variety of non-image forming physiological functions, such as circadian photoentrainment, pupillary light reflex and masking responses to light. In view of the recently proposed role of retinal inputs for the regulation of diurnal and nocturnal behavior, we performed the first deep analysis of the ipRGC system in a diurnal rodent model, Arvicanthis ansorgei , and compared the anatomical and physiological properties of ipRGCs with those of nocturnal mice. Based on somata location, stratification pattern and melanopsin expression, we identified two main ipRGC types in the retina of Arvicanthis : M1, constituting 74% of all ipRGCs and non-M1 (consisting mainly of the M2 type) constituting the following 25%. The displaced ipRGCs were rarely encountered. Phenotypical staining patterns of ganglion cell markers showed a preferential expression of Brn3 and neurofilaments in non-M1 ipRGCs. In general, the anatomical properties and molecular phenotyping of ipRGCs in Arvicanthis resemble ipRGCs of the mouse retina, however the percentage of M1 cells is considerably higher in the diurnal animal. Multi-electrode array recordings (MEA) identified in newborn retinas of Arvicanthis three response types of ipRGCs (type I, II and III) which are distinguished by their light sensitivity, response strength, latency and duration. Type I ipRGCs exhibited a high sensitivity to short light flashes and showed, contrary to mouse type I ipRGCs, robust light responses to 10 ms flashes. The morphological, molecular and physiological analysis reveals very few differences between mouse and Arvicanthis ipRGCs. These data imply that the influence of retinal inputs in defining the temporal niche could be related to a stronger cone input into ipRGCs in the cone-rich Arvicanthis retina, and to the higher sensitivity of type I ipRGCs and elevated proportion of M1 cells.

Aging-like circadian disturbances in folate-deficient mice

The elderly population shows various circadian disturbances, including dampened amplitude of rhythmicity and decreased responsiveness to light. The common poor folate status in the elderly might account for these aging-related circadian disturbances. To test this hypothesis, we investigated whether folate deficiency in mice affects circadian oscillations of the master clock in the suprachiasmatic nuclei, and the shifting responses to light. Mice fed a diet without folate for 6 weeks displayed markedly reduced (4.5-fold) erythrocyte folate concentration and increased (2.3-fold) homocysteinemia compared with control mice. Folate deficiency decreased the circadian amplitude of vasopressin and the clock protein PERIOD 2 (PER2) in the master clock, slowed the rate of re-entrainment of behavioral rhythms after delayed light-dark cycle and reduced light-induced phase-delays, without detectable morphologic changes in the retina, such as the number of melanopsinergic ganglion cells, that might have impaired photodetection. In conclusion, folate deficiency and consecutive hyperhomocysteinemia led to dampened PER2 and vasopressin oscillations in the master clock and reduced responsiveness to photic resetting, which constitute hallmarks of aging effects on circadian rhythmicity.

Oxygen-induced retinopathy induces short-term glial stress and long-term impairment of photoentrainment in mice

BackgroundRetinopathy of prematurity is a serious potentially blinding disease of pre-term infants. There is extensive vascular remodeling and tissue stress, but data concerning alterations in retinal neurons and glia, and long-term functional sequelae are still incomplete.MethodsROP was induced using the oxygen-induced retinopathy (OIR) mouse model. Postnatal day 7 (P7) 129SVE mice were exposed to hyperoxia (75u2009±u20090.5xa0% oxygen) for 5 days, and then returned to normoxia to induce OIR. Exposed animals were euthanized at 5 (P17-OIR) and 14 days (P26-OIR) after return to normal air, together with corresponding age-matched control mice (P17-C and P26-C respectively) raised only in room air. Their retinas were examined by immunohistochemistry using a battery of antibodies against key glial and neuronal proteins. A further group of OIR mice and controls were examined at 10 weeks of age for their ability to re-entrain to changing 12xa0h light/12xa0h dark cycles, assayed by wheel-running actimetry. In this protocol, animals were subjected to three successive conditions of 300 lux, 15 lux and 1 lux ambient light intensity coupled with 6 hours of jetlag. Animals were euthanized at 4 months of age and used in immunoblotting for rhodopsin.ResultsCompared to P17-C, immunohistochemical staining of P17-OIR sections showed up-regulation of stress-related and glutamate-regulatory proteins in astrocytes and Müller glial cells. In contrast, glial phenotypic expression in P26-OIR retinas largely resembled that in P26-C. There was no loss in total retinal ganglion cells (RGC) at either P17-OIR or P26-OIR compared to corresponding controls, whereas intrinsically photosensitive RGC showed significant decreases, with 375u2009±u200913/field in P26-OIR compared to 443u2009±u200930/field in P26-C (pu2009<u20090.05). Wheel actimetry performed on control and OIR-treated mice at 4 months demonstrated that animals raised in hyperoxic conditions had impaired photoentrainment at low illuminance of 1 lux, as well as significantly reduced levels of rhodopsin compared to age-matched controls.ConclusionsOIR leads to transient up-regulation of retinal glial proteins involved in metabolism, and partial degeneration of intrinsically photosensitive RGC and rod photoreceptors. OIR affects circadian photo-entrainment at low illuminance values, possibly by affecting the rod pathway and/or intrinsically photosensitive RGC input to the circadian clock. This study hence shows that retinopathy of prematurity affects light-regulated circadian behavior in an animal model, and may induce similar problems in humans.

Structural and physiological responses to prolonged constant lighting in the cone-rich retina of Arvicanthis ansorgei

Cone photoreceptor death is a leading cause of blindness in industrialised countries. Despite this, there are few mammalian models available to study cone pathophysiology. The diurnal rodent Arvicanthis possesses a high cone percentage and ease of maintenance. We recently described the effect of ambient light conditions on cyclic disc shedding, and observed that 24 h of constant illumination (LL) completely disrupted the normal rhythmic process and increased cone shedding fourfold. The current study was undertaken to see whether protracted constant illumination (7 days LL) would further perturb cone (and rod) turnover, and possibly lead to photoreceptor degeneration. Whereas control (cyclic lighting) retinas exhibited a typical early morning burst in phagosomes, LL retinas exhibited only low uniform numbers of rod and cone phagosomes across 24 h, with no peak of shedding at any time. Morphometric and immunohistochemical analyses of Arvicanthis retinas after 7 days LL (300 lux) showed no structural changes compared to control retinas.