September 2010
Volume 51, Issue 9
Free
Retinal Cell Biology  |   September 2010
Environmental Damage to the Retina and Preconditioning: Contrasting Effects of Light and Hyperoxic Stress
Author Affiliations & Notes
  • Yuan Zhu
    From the ARC Centre of Excellence in Vision Science and
    Division of Biomedical Sciences and Biochemistry, Research School of Biology, Australian National University, Canberra, ACT, Australia; and
    Save Sight Institute and Discipline of Physiology, The University of Sydney, Sydney, NSW, Australia.
  • Krisztina Valter
    From the ARC Centre of Excellence in Vision Science and
    Division of Biomedical Sciences and Biochemistry, Research School of Biology, Australian National University, Canberra, ACT, Australia; and
  • Jonathan Stone
    From the ARC Centre of Excellence in Vision Science and
    Save Sight Institute and Discipline of Physiology, The University of Sydney, Sydney, NSW, Australia.
  • Corresponding author: Yuan Zhu, GPO Box 475, Canberra, ACT 2601, Australia; [email protected]
Investigative Ophthalmology & Visual Science September 2010, Vol.51, 4821-4830. doi:https://doi.org/10.1167/iovs.09-5050
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Yuan Zhu, Krisztina Valter, Jonathan Stone; Environmental Damage to the Retina and Preconditioning: Contrasting Effects of Light and Hyperoxic Stress. Invest. Ophthalmol. Vis. Sci. 2010;51(9):4821-4830. https://doi.org/10.1167/iovs.09-5050.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose.: Environmental stress (bright light, hypoxia) can “condition” retinal photoreceptors, increasing their resistance to subsequent stress. The present study tests whether another photoreceptor-lethal stress, hyperoxia, can induce similar resistance.

Methods.: Vulnerability to hyperoxia was tested in young adult C57BL/6J mice exposed to 1000 lux cyclic light for 1 week or to 50% O2 for 1 week and then to 75% O2 for 2 weeks. Vulnerability to light was tested in Balb/cJ mice exposed to 300 lux cyclic light for 2 days or to 75% O2 for 2 weeks and then to 1000 lux cyclic light for 1 week. Retinas were analyzed for photoreceptor death, levels of stress-related proteins (GFAP, FGF-2, MnSOD, acrolein), and the regulation of candidate neuroprotective genes (HSP70.1, Ledgf, FGF-13, Timp2).

Results.: Light preconditioning did not cause measurable death of photoreceptors but reduced photoreceptor death induced by subsequent hyperoxic or light stress, reduced levels of stress-related proteins, and maintained the length and organization of photoreceptor outer segments. Hyperoxic preconditioning caused measurable cell death but provided no protection against subsequent hyperoxic or light stress. Of the four candidate neuroprotective proteins examined, the regulation of only one (Timp2) seemed associated with the neuroprotection observed.

Conclusions.: Light preconditioning, causing only minimal damage to photoreceptors, induced protection against subsequent stress from both hyperoxia and light. By contrast, hyperoxic preconditioning caused measurable photoreceptor damage but induced no protection against light or hyperoxia. These data suggest a separation between stress-induced damage to photoreceptors and the upregulation of protective mechanisms, encouraging the search for ways to protect the retina without damaging it.

The stability of photoreceptors in the adult retina is significantly determined by environmental factors, especially the retina's previous exposure to light. Separately from the adaptation of the phototransduction cascade to light, the photoreceptor cell adapts to raised levels of illumination, within the physiological range, by regulating the length of its outer segment (OS), 14 its expression of stress-inducible factors, 5,6 and the complexity of the synapses it forms in the outer plexiform layer. 79 One result of these changes is that the retina, “conditioned” by the experience of light, is more resistant to acute light damage. 
Several questions arise from the concept of conditioning, sometimes called preconditioning. One question is teleological: Why have mechanisms of protection evolved that are light inducible? Why are such mechanisms not switched on all the time? The answer to this question appears to be that some of these mechanisms reduce the sensitivity of the photoreceptor to light. This reduction is best documented for fibroblast growth factor-2 (FGF-2). 1012 Another question is whether the mechanisms of preconditioning are elicited specifically by light or can be elicited by other environmental factors that are stressful to the retina, such as hypoxia and hyperoxia. Hypoxia and ischemia have also been shown to elicit a preconditioning response, mediated by erythropoietin. 1317 Yet another question is whether some level of photoreceptor damage by the preconditioning stress is necessary for the protective effect. 
This study examined whether hyperoxia can precondition the retina, testing whether preconditioning is mediated by known stress-related proteins, and whether retinal damage is necessary for the induction of protective conditioning. The question is of interest clinically because the outer retina becomes hyperoxic during the course of retinal degeneration as the photoreceptor population is depleted and its consumption of oxygen is reduced (for a review, see Ref. 18). The question then arises whether this chronic hyperoxia contributes to the progression of the degeneration because oxygen is toxic to photoreceptors 1922 or, conversely, whether hyperoxia evokes a self-protective response from the retina that might stabilize the degenerating retina. 
We have, therefore, tested whether light conditioning protects against hyperoxic and light stress and whether hyperoxic conditioning protects against either light or hyperoxia. Results indicate that light preconditioning can protect the retina against subsequent stress from either light or hyperoxic stress, whereas hyperoxic preconditioning protects against neither. 
Materials and Methods
Mouse Strains and General Handling
Observations were made in young adult (aged postnatal day [P] 70–180) mice from two strains (Fig. 1); the observations on photoreceptor death, retinal morphology, and immunohistochemistry were made on animals aged P70 to P120; quantitative PCR included animals aged P180. Walsh et al. 23 demonstrated a consistent difference in the vulnerability of retinal photoreceptors between the Balb/cJ strain, in which photoreceptors are vulnerable to light damage but resistant to hyperoxic stress, and the C57BL/6J strain, in which photoreceptors are resistant to light damage and vulnerable to hyperoxia. Protection against light-induced damage was, therefore, tested in the Balb/cJ strain and against hyperoxia-induced damage in the C57BL/6J strain. The mice were born and raised in room air and dim cyclic illumination (12 hours 5 lux/12 hours dark). Standard rodent chow and water were provided ad libitum. Euthanatization was performed with pentobarbitone sodium (Lethabarb, 60 mg/kg administered intraperitoneally; Virbac Animal Health Pty., Milperra, NSW, Australia). All experimental procedures were in compliance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and the requirements of The Australian National University Animal Experimentation Ethics Committee. 
Figure 1.
 
Experimental design. Two strains of mice were studied: C57BL/6J because their photoreceptors are hyperoxia vulnerable and BALB/cJ because their photoreceptors are light vulnerable. In each strain, three groups were studied: control, light-preconditioned, and hyperoxia-preconditioned. For the C57BL/6J strain, the test stress was hyperoxia; for the BALB/cJ strain, the test stress was bright light.
Figure 1.
 
Experimental design. Two strains of mice were studied: C57BL/6J because their photoreceptors are hyperoxia vulnerable and BALB/cJ because their photoreceptors are light vulnerable. In each strain, three groups were studied: control, light-preconditioned, and hyperoxia-preconditioned. For the C57BL/6J strain, the test stress was hyperoxia; for the BALB/cJ strain, the test stress was bright light.
Light and Hyperoxic Preconditioning
Animals raised in dim cyclic light served as controls. C57BL/6J mice were studied in three groups—one maintained in control conditions, one exposed to bright cyclic light (1000 lux) for 1 week, and one exposed to 50% oxygen for 1 week—after which all three groups were exposed to 75% O2 for 2 weeks (test stress). The Balb/cJ mice were also studied in three groups—one kept in control conditions, one exposed to 300 lux cyclic illumination for 2 days, and one exposed to 75% O2 for 2 weeks—after which all were exposed to bright light stressed with bright (1000 lux) cyclic light for 1 week (Fig. 1). Eyes were harvested at the end of each test stress period. Control eyes were harvested at age-matched times. The group size for each experimental condition was five. The hyperoxic conditioning experiments were repeated, however, giving larger numbers at those experimental conditions. 
For the BALB/cJ strain (which is oxygen resistant), the conditioning exposure to oxygen was set at a relatively high level (75% for 2 weeks) to ensure that some photoreceptor damage occurred. This was done to ensure that the observed lack of a hyperoxia-induced protective conditioning was not caused by the mildness of the hyperoxic preconditioning. 
For light exposure, animals were kept individually in polymethyl methacrylate compartments with a white fluorescent light 10 cm above the cage. The spectral distribution of light source was uniform over the visible spectrum. For oxygen exposure, animals in their normal cages were placed in polymethyl methacrylate chambers, in which the oxygen level was controlled by a feedback device (Oxycycler; Biospherix, Lacona, NY). 
Preparation of Material
Cryosectioning.
The superior aspect of the eye was marked with an insoluble projection pen, and eyes were fixed by immersion in 4% paraformaldehyde fixative buffer at 4°C for 3 hours. After three rinses in 0.1 M phosphate-buffered saline (PBS), eyes were left overnight in a 15% sucrose solution to provide cryoprotection. Eyes were embedded in mounting medium (Tissue-Tek OCT compound; Sakura Finetek, Torrance, CA) by snap freezing in liquid nitrogen. Cryosections were cut at 16 μm (CM1850 Cryostat; Leica, Wetzlar, Germany) through the entire retina along the vertical meridian of the eye to produce sections of the full extent of the eye from superior to inferior edge, which were mounted on gelatin and poly-l-lysine–coated slides. Sections were then dried overnight in a 50°C oven and stored at −20°C until processed. 
Quantitative Real-Time PCR.
Each retina was dissected from the eyecup after removal of the cornea and lens. Both retinas of each animal were pooled. Total retinal RNA extraction was performed using a combination of reagent (TRIzol; Invitrogen Life Technologies, Carlsbad, CA) and a micro kit (RNAqueous; Ambion, Foster City, CA). The reagent was used to isolate the RNA, and the kit was used to purify and DNase-treat the RNA. Each retina was quickly placed into a 1.5-mL tube containing 200 μL reagent. After the homogenization of retinas on ice, another 660 μL reagent (TRIzol; Invitrogen Life Technologies) and 160 μL chloroform were added to the tube. The tube was vortexed for 20 seconds and allowed to stand for 7 minutes at room temperature. The tubes were centrifuged at 13,000g for 10 minutes at 4°C. The supernatant was then removed and placed into a clean 1.5-mL tube with half its volume of 100% ethanol. The tube was vortexed briefly before contents underwent purification and DNase treatment, as detailed in the micro kit manual ((RNAqueous; Ambion). Purified DNAse-treated RNA was analyzed on a spectrophotometer (ND-1000; NanoDrop Technologies, Wilmington, DE) and a bioanalyzer (2100 Bioanalyzer; Agilent Technologies, Santa Clara, CA) to determine the quantity and quality of the sample. First-strand cDNA synthesis was achieved by reverse transcribing 1 μg total RNA using the reverse transcriptase protocol (Superscript III; Invitrogen Life Technologies, Carlsbad, CA). A cocktail of 1 μL oligo primer, 1 μL 10 mM dNTP, 4 μL 5× RT buffer, 1 μL 0.1 M dithiothreitol, 1 μL 40 U/μL ribonuclease inhibitor (RNaseOut; Invitrogen Life Technologies), and 1 μL reverse transcriptase (Superscript III; Invitrogen Life Technologies) was added to 1 μg retinal RNA and made up to 20 μL with RNase-free water. The mixture was incubated at 50°C for 1 hour before the reaction was terminated by increasing the temperature to 70°C for 15 minutes. 
Detection of Cell Death (TUNEL)
Retinal sections were labeled for apoptotic cell death using the terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) technique, 24 according to protocols published previously. 25 To demonstrate cellular layers, sections were also labeled with the DNA-specific dye bisbenzimide (Calbiochem, La Jolla, CA), by incubation for 2 minutes in a 1:10,000 solution in 0.1 M PBS. Sections cut adjacent to or through the optic nerve head (ONH) were chosen to minimize variations in retinal length and position. Counts of TUNEL-positive profiles (apoptotic cells) were made using a calibrated 20× objective and an eyepiece graticule. Each section was scanned from the superior to the inferior edge, and the number of TUNEL-positive profiles was recorded for each 400-μm length of the section. Counts were averaged from at least four sections per animal and were recorded separately for the outer nuclear layer (ONL) and the inner nuclear layer (INL). 
Photoreceptor Survival (ONL/INL Ratio)
Sections through the ONH were stained with the DNA-specific dye bisbenzimide (1:10,000). The thicknesses of the ONL and the INL were measured on digital images at 0.24-mm intervals from the ONH toward the inferior and superior ora serrata. The ratio of ONL to INL thickness was calculated to compensate for oblique sectioning. For each experimental group, a single section from the retinas of at least six eyes was measured. 
Immunohistochemistry
Retinal sections were pretreated with 70% ethanol (15 minutes), washed with distilled H2O (5 minutes), washed twice with 0.1 M PBS (5 minutes each time), and incubated in 10% normal goat serum in 0.1 M PBS for 1 hour at room temperature to block nonspecific binding. Sections were then incubated overnight at 4°C with one of the following primary antibodies: rabbit polyclonal anti-GFAP (1:700; DakoCytomation, Campbellfield, Australia), mouse monoclonal anti-FGF2 (1:200; Upstate Biotechnology Inc., Lake Placid, NY), rabbit polyclonal anti-acrolein (1:500; Cell Sciences, Canton, MA), rabbit polyclonal anti-MnSOD (1:200; Upstate Biotechnology Inc.), and mouse anti-rhodopsin and polyclonal anti-rabbit L/M opsin (1:1000; Chemicon, Temecula, CA). After three rinses in PBS for 10 minutes each, sections were incubated with the appropriate secondary antibody for 1 hour at room temperature or overnight at 4°C before they were coverslipped with a glycerol-gelatin medium. Images were taken by confocal microscopy (LSM 5 Pascal; Carl Zeiss Microimaging GmbH, Oberkochen, Germany) and analyzed with ImageJ software (developed by Wayne Rasband, National Institutes of Health, Bethesda, MD; available at http://rsb.info.nih.gov/ij/index.html). During the acquisition of images from animals in different treatment groups, photomultiplier settings were held constant. Measurements of FGF-2 labeling were made across the full thickness of the ONL. Measurements of acrolein labeling were made over an area covering the thickness of the ONL, inner segment (IS), and OS. Measurements of MnSOD labeling were made across the most strongly MnSOD-labeled region in the retina, the IS. The program Image J 1.4 (National Institutes of Health) was used to quantify fluorescence intensities in captured images. 
Quantitative PCR
As a second approach to identifying genes important for the neuroprotective effect of preconditioning, we relied on the microarray study of Huang et al., 26 who demonstrated that Hsp70.1, Fgf13, Ledgf, and Timp2 are upregulated by increased ambient light and have been identified as neuroprotection related. We generated three groups of C57BL/6J mice: control, light (1000 lux for 1 week) preconditioned, and hyperoxia (50% for 1 week) preconditioned. Four biological replicates were used for each condition. Retinal RNA (1 μg) was reverse transcribed into cDNA using the reverse transcriptase (Superscript III; Invitrogen Life Technologies) protocol. The resultant cDNA sample served as a template for real time-quantitative PCR using TaqMan probes and the accompanying gene expression mix (Master Mix; Applied Biosystems, Foster City, CA). TaqMan gene expression assay identification numbers used included Mm01159846_s1 for Hsp70.1, Mm00438910_m1 for Fgf13, Mm00505918_m1 for Ledgf, Mm00441825_m1 for Timp2, and Mm00520345_m1 for Rho. A typical 20-μL reaction is composed of 10 μL 2× gene expression mix (Master Mix; Applied Biosystems), 1 μL TaqMan gene expression assay, 0.4 μL cDNA templates, and RNase-free water. All reactions were set up in duplicate, and negative controls with no template and no probe were included. Quantitative PCR was performed (Rotor-Gene 3000; Mortlake, Australia) and analyzed using the appropriate software (Rotor-Gene 6; Corbett Robotics). The quantitative PCR program included 2 minutes' incubation at 50°C, 10 minutes' enzyme activation at 95°C followed by 40 cycles of 15 seconds' denaturation at 95°C, and 1-minute probe annealing at 60°C. The cycle threshold means, representing the quantitation of the amount of cDNA in the original sample, were used to calculate fold-change using the Pfaffl equation. 27 GADH was used as a reference gene relative to which values of other genes were calculated. 
Statistical Analysis
All values are mean ± SEM. A two tailed unpaired t-test was used to test differences between groups. P < 0.05 was considered statistically significant. 
Results
Photoreceptor Vulnerabilities in the Balb/cJ and C57BL/6J Strains
When C57BL/6J mice were exposed to 75% oxygen for 2 weeks, the frequency of TUNEL-positive (dying) cells in the ONL increased significantly above control levels, confirming previous works, 2023 and the increase was almost entirely restricted to the ONL (Figs. 2A, 2B) and was most prominent in the inferior retina close to the optic disc. That is, hyperoxia was specifically toxic to photoreceptors. The same oxygen exposure induced more limited photoreceptor death in the Balb/cJ mouse (Figs. 2D, 2E), confirming previous work. 23 When Balb/cJ mice were exposed to 1000 lux light for 1 week, the frequency of TUNEL-positive (dying) cells in the ONL increased significantly above control levels (Figs. 2D, 2F), and the increase in cell death was almost entirely restricted to the ONL. In other words, light was specifically toxic to photoreceptors. The same light exposure did not induce a measurable increase in photoreceptor death in the C57BL/6J mouse (Figs. 2A, 2C), confirming previous work. 28  
Figure 2.
 
Sections of mouse retina labeled for dying cell (red) with the TUNEL technique and for surviving cells (blue) with bisbenzimide. (A, D) Control retinas of C57BL/6J and Balb/cJ strains showed few TUNEL-positive cells. (B, E) Two weeks' exposure to 75% hyperoxia caused a large increase in the number of TUNEL-positive cells in the ONL of the C57BL/6J mouse and a lesser increase in the Balb/cJ mouse. The ONL was thinned in the C57BL/6J mouse. (C, F) One week's exposure to 1000 lux light caused no measurable rise in the number of TUNEL-positive cells in the C57BL/6J mouse but increased TUNEL-positive cell numbers in the ONL of the Balb/cJ mouse. The ONL was thinned by light exposure in the Balb/cJ mouse. Scale bar, 20 μm (A–F).
Figure 2.
 
Sections of mouse retina labeled for dying cell (red) with the TUNEL technique and for surviving cells (blue) with bisbenzimide. (A, D) Control retinas of C57BL/6J and Balb/cJ strains showed few TUNEL-positive cells. (B, E) Two weeks' exposure to 75% hyperoxia caused a large increase in the number of TUNEL-positive cells in the ONL of the C57BL/6J mouse and a lesser increase in the Balb/cJ mouse. The ONL was thinned in the C57BL/6J mouse. (C, F) One week's exposure to 1000 lux light caused no measurable rise in the number of TUNEL-positive cells in the C57BL/6J mouse but increased TUNEL-positive cell numbers in the ONL of the Balb/cJ mouse. The ONL was thinned by light exposure in the Balb/cJ mouse. Scale bar, 20 μm (A–F).
Preconditioning with Light
Effect on Light-Induced Cell Death (Balb/cJ).
Previous studies have established the effectiveness of exposure of the retina to raised ambient light, 29,30 or to briefer periods of more intense light, in reducing the vulnerability of photoreceptors in the rat and mouse to light damage. 5,31,32 It is also well established that exposure to hypoxia 1617,33,34 and ischemia 13,35,36 reduces photoreceptor vulnerability to light damage in rat and mouse. The conditioning effect of light was confirmed here. The preconditioning stress (300 lux for 2 days) did not measurably increase photoreceptor death over control levels (Figs. 3A, 3B, 3E). The test stress caused a clear increase in the rate of cell death (Figs. 2A, 2D) and measurable thinning of the ONL (Fig. 3F). The death of photoreceptors caused by 1000 lux light exposure for 1 week was reduced by preconditioning (Figs. 3C, 3D, 3E). Preconditioning also reduced the thinning of the ONL caused by the 1000 lux 1-week exposure (Fig. 3F), maintaining ONL thickness at near control levels. 
Figure 3.
 
Effects of light preconditioning on light-induced cell death in Balb/cJ retina. Four groups of animals were analyzed (each group contained five animals). (A–D) TUNEL labeling (red) shows dying cells in the sensitive superior region of the retina of Balb/cJ mice after different treatment. The nuclei of retinal cells are labeled blue with bisbenzimide. The number of TUNEL-positive profiles was not raised by 2 days' exposure to 300 lux (A, B) but was markedly raised by 1 week's exposure to 1000 lux (A, C). The light-induced rise was reduced by light preconditioning (D). (E) Mean numbers of TUNEL-positive profiles/section in the ONL of control and treated Balb/cJ retinas. Each symbol represents the mean for one animal; the horizontal line marks the mean of each group; the error bars represent ± SEM. Preconditioning (300 lux for 2 days) did not increase TUNEL labeling; the test stress (1000 lux for 1 week) produced a very significant increase. This increase was reduced by preconditioning, and the reduction was statistically significant (P = 0.001). (F) ONL thickness, estimated by the ONL/INL ratio, plotted as a function of distance from the optic nerve head (ONH) along the vertical meridian of the eye. The mean value for each group is plotted, with error bars representing the SEM of measures at each point. After 1-week exposure of damaging light, the ONL thickness was reduced in the unconditioned retina, most markedly just superior to the optic disc. The reduction was decreased by preconditioning. Asterisks: points at which the ONL thickness in the conditioned light-damaged group (300 lux for 2 days, then 1000 lux for 1 week) was significantly (P < 0.05 on a 2-tailed t-test) greater than in the light-damaged group (1000 lux for 1 week). That is, these are the points at which the rescue of the ONL reached statistical significance. Scale bar, 20 μm (A–D).
Figure 3.
 
Effects of light preconditioning on light-induced cell death in Balb/cJ retina. Four groups of animals were analyzed (each group contained five animals). (A–D) TUNEL labeling (red) shows dying cells in the sensitive superior region of the retina of Balb/cJ mice after different treatment. The nuclei of retinal cells are labeled blue with bisbenzimide. The number of TUNEL-positive profiles was not raised by 2 days' exposure to 300 lux (A, B) but was markedly raised by 1 week's exposure to 1000 lux (A, C). The light-induced rise was reduced by light preconditioning (D). (E) Mean numbers of TUNEL-positive profiles/section in the ONL of control and treated Balb/cJ retinas. Each symbol represents the mean for one animal; the horizontal line marks the mean of each group; the error bars represent ± SEM. Preconditioning (300 lux for 2 days) did not increase TUNEL labeling; the test stress (1000 lux for 1 week) produced a very significant increase. This increase was reduced by preconditioning, and the reduction was statistically significant (P = 0.001). (F) ONL thickness, estimated by the ONL/INL ratio, plotted as a function of distance from the optic nerve head (ONH) along the vertical meridian of the eye. The mean value for each group is plotted, with error bars representing the SEM of measures at each point. After 1-week exposure of damaging light, the ONL thickness was reduced in the unconditioned retina, most markedly just superior to the optic disc. The reduction was decreased by preconditioning. Asterisks: points at which the ONL thickness in the conditioned light-damaged group (300 lux for 2 days, then 1000 lux for 1 week) was significantly (P < 0.05 on a 2-tailed t-test) greater than in the light-damaged group (1000 lux for 1 week). That is, these are the points at which the rescue of the ONL reached statistical significance. Scale bar, 20 μm (A–D).
Effect on Hyperoxia-Induced Cell Death (C57BL/6J).
This effect was tested in a protocol similar to that used in Figure 3; the test stress was hyperoxia (rather than light), and the C57BL/6J strain was used because it is vulnerable to hyperoxia. The preconditioning stress (1000 lux for 1 week) did not measurably increase photoreceptor death over control levels (Figs. 4A, 4B, 4E). The death rate of photoreceptors caused by exposure to 75% oxygen for 2 weeks was reduced by preconditioning (Figs. 4C-E). Preconditioning also reduced the thinning of the ONL caused by test stress (Fig. 4F). 
Figure 4.
 
Effects of light preconditioning on hyperoxia-induced cell death in C57BL/6J retina. Four groups of animals were analyzed (each group contained six to nine animals). (A–D) Representative pictures of TUNEL staining in the inferior retina of C57BL/6J mice after different treatment. The number of TUNEL-positive profiles was not raised by 1 week's exposure to 1000 lux (A, B) and was markedly raised by 2 weeks' exposure to 75% O2 (A, C). The hyperoxia-induced increase was reduced by previous exposure to light (D). (E) Mean number of TUNEL-positive profiles/section in the ONL of control and treated C57BL/6J retinas. Each symbol represents the mean for one animal, the horizontal line marks the mean of each group, and the error bars represent ± SEM. Preconditioning (1000 lux for 1 week) did not increase TUNEL labeling; the test stress (75% oxygen for 2 weeks) produced a very significant increase. This increase was reduced by preconditioning, and the reduction was statistically significant (P < 0.01). (F) ONL thickness, estimated by the ONL/INL ratio, plotted as a function of distance from the optic nerve head (ONH) along the vertical meridian of the eye. The mean value for each group is plotted, with error bars representing the SEM of measures at each point. After 2 weeks' exposure to 75% oxygen, the ONL thickness was reduced in the unconditioned retina, most markedly just inferior to the optic disc. Asterisks: points at which the thickness of the ONL in the conditioned hyperoxia-damaged group (1000 lux for 1 week, then 75% O2 for 2 weeks) was significantly (P < 0.05 on a 2-tailed t-test) greater than in the unconditioned hyperoxia-damaged group (75% O2 for 2 weeks). That is, these are the points at which the rescue of the ONL reached statistical significance. Scale bar, 20 μm (A–D).
Figure 4.
 
Effects of light preconditioning on hyperoxia-induced cell death in C57BL/6J retina. Four groups of animals were analyzed (each group contained six to nine animals). (A–D) Representative pictures of TUNEL staining in the inferior retina of C57BL/6J mice after different treatment. The number of TUNEL-positive profiles was not raised by 1 week's exposure to 1000 lux (A, B) and was markedly raised by 2 weeks' exposure to 75% O2 (A, C). The hyperoxia-induced increase was reduced by previous exposure to light (D). (E) Mean number of TUNEL-positive profiles/section in the ONL of control and treated C57BL/6J retinas. Each symbol represents the mean for one animal, the horizontal line marks the mean of each group, and the error bars represent ± SEM. Preconditioning (1000 lux for 1 week) did not increase TUNEL labeling; the test stress (75% oxygen for 2 weeks) produced a very significant increase. This increase was reduced by preconditioning, and the reduction was statistically significant (P < 0.01). (F) ONL thickness, estimated by the ONL/INL ratio, plotted as a function of distance from the optic nerve head (ONH) along the vertical meridian of the eye. The mean value for each group is plotted, with error bars representing the SEM of measures at each point. After 2 weeks' exposure to 75% oxygen, the ONL thickness was reduced in the unconditioned retina, most markedly just inferior to the optic disc. Asterisks: points at which the thickness of the ONL in the conditioned hyperoxia-damaged group (1000 lux for 1 week, then 75% O2 for 2 weeks) was significantly (P < 0.05 on a 2-tailed t-test) greater than in the unconditioned hyperoxia-damaged group (75% O2 for 2 weeks). That is, these are the points at which the rescue of the ONL reached statistical significance. Scale bar, 20 μm (A–D).
Effect on Expression of Stress-Related Proteins and Structure of Outer Segments.
Glial fibrillary acidic protein (GFAP) is an intermediate filament protein normally expressed in astrocytes at the inner surface of the retina, not in the radially oriented processes of Müller cells (Fig. 5A). In the C57BL/6J mouse, the preconditioning stress (1000 lux for 1 week) did not change this pattern of expression (Fig. 5B). The test stress (75% oxygen for 2 weeks) upregulated GFAP expression in Müller cells, most prominently in the inferior retina, where photoreceptor death was maximal (compare Figs. 5C, 5D). This upregulation was reduced when the test stress was preceded by the preconditioning stress (Figs. 5E, 5F). FGF-2 is a stress-inducible trophic factor normally present at low levels in the somas of photoreceptors but prominent under stress. 37,38 FGF-2 was also upregulated by the test stress, most prominently in the inferior retina (Figs. 5C, 5D); this upregulation was reduced by preconditioning (Figs. 5E, 5F). 
Figure 5.
 
Effects of light preconditioning on the expression of stress-related proteins (A–F) and on OS length (K–N) in the C57BL/6J retina. (A–F) GFAP (red) is confined to astrocytes at the inner surface of the control retina (A). After 2 weeks' hyperoxia, its expression was upregulated in Müller cells in the inferior retina (D). This reduction was decreased by light preconditioning (F). FGF-2 (green) was not prominent in the ONL in control retinas but was upregulated after 2 weeks' hyperoxia (in inferior retina, D). This upregulation was reduced by light preconditioning (F). (G–J) Acrolein labeling in the vulnerable inferior region was upregulated after 2 weeks' hyperoxia (I). The upregulation was reduced by light preconditioning (J). (K–N) MnSOD labeling in the vulnerable region of inferior retina. Labeling of IS was upregulated after 2 weeks' hyperoxia; the upregulation was reduced by light preconditioning. (O–R) Labeling for rhodopsin (red) label and L/M opsin (green) in the vulnerable region of the inferior retina indicates that the lengths of rod and cone OS were reduced by 2 weeks' hyperoxia; the reduction was mitigated by light preconditioning. GCL, ganglion cell layer; RPE, retinal pigment epithelium. Scale bars: 20 μm (A–N, O–R).
Figure 5.
 
Effects of light preconditioning on the expression of stress-related proteins (A–F) and on OS length (K–N) in the C57BL/6J retina. (A–F) GFAP (red) is confined to astrocytes at the inner surface of the control retina (A). After 2 weeks' hyperoxia, its expression was upregulated in Müller cells in the inferior retina (D). This reduction was decreased by light preconditioning (F). FGF-2 (green) was not prominent in the ONL in control retinas but was upregulated after 2 weeks' hyperoxia (in inferior retina, D). This upregulation was reduced by light preconditioning (F). (G–J) Acrolein labeling in the vulnerable inferior region was upregulated after 2 weeks' hyperoxia (I). The upregulation was reduced by light preconditioning (J). (K–N) MnSOD labeling in the vulnerable region of inferior retina. Labeling of IS was upregulated after 2 weeks' hyperoxia; the upregulation was reduced by light preconditioning. (O–R) Labeling for rhodopsin (red) label and L/M opsin (green) in the vulnerable region of the inferior retina indicates that the lengths of rod and cone OS were reduced by 2 weeks' hyperoxia; the reduction was mitigated by light preconditioning. GCL, ganglion cell layer; RPE, retinal pigment epithelium. Scale bars: 20 μm (A–N, O–R).
Acrolein is a breakdown product of oxidative damage to lipids. Its appearance in the C57BL/6J retina was slightly increased by the preconditioning stress (Figs. 5G, 5H). Its prominence was markedly increased by the test stress in photoreceptor cells (OS, IS, ONL) (Fig. 5I), and this increase was reduced by the preconditioning (Fig. 5J). 
MnSOD is the mitochondrial form of superoxide dismutase, a free radical scavenger, and is upregulated in response to mitochondrial damage. 39,40 Its expression in the mitochondria of IS in the C57BL/6J retina was not affected by the preconditioning stress (Figs. 5K, 5L). The test stress upregulated MnSOD specifically in IS in inferior retina (Fig. 5M); this increase was reduced by preconditioning (Fig. 5N). 
In the C57BL/6J retina, the preconditioning stress (1000 lux for 1 week) did not measurably reduce OS length (Figs. 5O, 5P). The test stress (75% oxygen for 2 weeks) caused a marked reduction in the length of both rod (red) and cone (green) OS and the appearance of vacuoles in the layer of OS in inferior retina (Fig. 5Q). Preconditioning with light prevented the formation of vacuoles and maintained OS length close to control level (Fig. 5R). 
Preconditioning with Hyperoxia
Effects on Light- and Hyperoxia-Induced Cell Death.
We then tested whether preconditioning with hyperoxia can reduce the photoreceptor death induced in the C57BL/6J mice by hyperoxia, and in the Balb/cJ mice by bright light. Exposure of C57BL/6J mice to 50% oxygen for 1 week increased photoreceptor death by a small but significant amount (Fig. 6A; control vs. 50% O2). This exposure did not reduce the photoreceptor death caused by a subsequent 2 weeks' exposure to 75% oxygen (75% O2 for 2 weeks vs. 50% O2 for 1 week, then 75% O2 for 2 weeks). In Balb/cJ mice, exposure to 75% oxygen for 2 weeks caused a limited but significant rise in photoreceptor death (Fig. 6B; control vs. 75% O2 for 2 weeks). When this exposure was followed by exposure to damaging light, the photoreceptor death induced by the damaging light was not reduced (1000 lux 1 week vs. 75% O2, then 1000 lux for 1 week). Correspondingly, there was also no evidence of preservation of the ONL (data not shown). In neither strain could we demonstrate that previous exposure to hyperoxia preconditioned the retina in a way that reduced photoreceptor vulnerability. 
Figure 6.
 
Effects of hyperoxic preconditioning on hyperoxia-induced photoreceptor death in the C57BL/6J mouse (A) and on light- induced photoreceptor death in the Balb/cJ mouse (B). Four groups of animals were examined for each series (each group contained five animals). Mean values and SEMs are shown for each group. Asterisks: the increased frequency of TUNEL-positive cells caused by hyperoxic preconditioning was significantly (*P < 0.001) greater than in control. Hyperoxic preconditioning did not cause a reduction in the frequency of TUNEL labeling induced by either hyperoxia (A) or light (B) test stress.
Figure 6.
 
Effects of hyperoxic preconditioning on hyperoxia-induced photoreceptor death in the C57BL/6J mouse (A) and on light- induced photoreceptor death in the Balb/cJ mouse (B). Four groups of animals were examined for each series (each group contained five animals). Mean values and SEMs are shown for each group. Asterisks: the increased frequency of TUNEL-positive cells caused by hyperoxic preconditioning was significantly (*P < 0.001) greater than in control. Hyperoxic preconditioning did not cause a reduction in the frequency of TUNEL labeling induced by either hyperoxia (A) or light (B) test stress.
Effects on Expression of Stress-Related Proteins.
In C57BL/6J mice, hyperoxic preconditioning (50% O2 for 1 week) did not cause a detectable change in the expression of GFAP or FGF-2 in the C57BL/6J retina (Figs. 7A, 7B), even though it did measurably raise the rate of photoreceptor death (Fig. 6). As already noted, a longer period of hyperoxia (75% O2 for 2 weeks) caused photoreceptor death in the inferior retina and upregulated the expression of both GFAP and FGF-2. Preconditioning with hyperoxia reduced the level of expression of FGF-2 but not of GFAP (Figs. 7D, 7F). 
Figure 7.
 
Effects of hyperoxia preconditioning on retinal expression of the stress-related proteins GFAP (red) and FGF-2 (green). (A–F) 75% hyperoxia for 2 weeks in the C57BL/6J retina upregulated the expression of GFAP in Müller cells (red) and of FGF-2 (green) in the ONL inferior retina (F). The upregulation of FGF-2, but not of GFAP, appeared to be mitigated by previous exposure to hyperoxia. (G–J) Light exposure of the Balb/cJ retina caused a limited upregulation of GFAP in Müller cells, in both the superior and the inferior retina (I) that was more prominent after hyperoxia preconditioning (J). FGF-2 expression in the ONL was also most prominent after the retina was exposed to both hyperoxia preconditioning and light stress (J). Scale bar, 20 μm (A–J).
Figure 7.
 
Effects of hyperoxia preconditioning on retinal expression of the stress-related proteins GFAP (red) and FGF-2 (green). (A–F) 75% hyperoxia for 2 weeks in the C57BL/6J retina upregulated the expression of GFAP in Müller cells (red) and of FGF-2 (green) in the ONL inferior retina (F). The upregulation of FGF-2, but not of GFAP, appeared to be mitigated by previous exposure to hyperoxia. (G–J) Light exposure of the Balb/cJ retina caused a limited upregulation of GFAP in Müller cells, in both the superior and the inferior retina (I) that was more prominent after hyperoxia preconditioning (J). FGF-2 expression in the ONL was also most prominent after the retina was exposed to both hyperoxia preconditioning and light stress (J). Scale bar, 20 μm (A–J).
In Balb/cJ mice, preconditioning with 75% oxygen for 2 weeks, which raised the rate of photoreceptor death measurably (Fig. 6), caused a limited upregulation of GFAP expression in Müller cells (Figs. 7G, 7H). The upregulation of GFAP expression caused by damaging light (1000 lux for 1 week; Fig. 7I) was not reduced by preconditioning with hyperoxia. Indeed hyperoxia appeared to upregulate the expression of both GFAP and FGF-2 (Fig. 7J). 
Regulation of Neuroprotective Genes by Preconditioning Stimuli
To explore the mechanisms underlying the protective effect of preconditioning in C57BL/6J photoreceptors, the expression of five genes—four potential neuroprotective genes and one nonregulated, retina-relevant gene—was examined in light-conditioned (1000 lux for 1 week) and hyperoxia-conditioned (50% O2 for 1 week) retinas using quantitative PCR (Fig. 8). Both light and hyperoxia preconditioning caused modest (1.34- to 1.53-fold) increases in the expression of Hsp70.1 and Fgf13. However, upregulation showed no significant difference between the two preconditioning stresses. The expression of Ledgf decreased slightly (to 0.88-fold) after light preconditioning and increased slightly after hyperoxia conditioning (1.27-fold), so that the difference in Ledgf expression between the two conditions appeared statistically significant. The most striking difference between light and hyperoxia conditioning occurred for Timp2, whose expression increased modestly (1.36-fold) after hyperoxia preconditioning but increased strongly (2.98-fold) after light preconditioning. 
Figure 8.
 
Quantitative real-time PCR analysis of potential neuroprotective genes in C57BL/6J retina after light and hyperoxia preconditioning. The expression level of genes in preconditioned groups was compared to that in the control group, and a value was generated for the fold-change in expression. A fold-change of 1 (dashed line) indicates no change in expression of the gene, a fold-change >1 indicates upregulation in the experimental group, and a fold-change <1 indicates downregulation. Four biological replicates were used in each precondition, with the error bar indicating the SEM. Hyperoxia caused only limited regulation of the four genes tested (Rho was selected as a retinally expressed gene whose expression is not known to be regulated by hyperoxia or light). Light preconditioning caused a significant upregulation of Timp2. Asterisks: regulation of the Ledgf and Timp2 was significantly different between light and hyperoxia conditioning (***P < 0.001 on a 2-tailed t-test).
Figure 8.
 
Quantitative real-time PCR analysis of potential neuroprotective genes in C57BL/6J retina after light and hyperoxia preconditioning. The expression level of genes in preconditioned groups was compared to that in the control group, and a value was generated for the fold-change in expression. A fold-change of 1 (dashed line) indicates no change in expression of the gene, a fold-change >1 indicates upregulation in the experimental group, and a fold-change <1 indicates downregulation. Four biological replicates were used in each precondition, with the error bar indicating the SEM. Hyperoxia caused only limited regulation of the four genes tested (Rho was selected as a retinally expressed gene whose expression is not known to be regulated by hyperoxia or light). Light preconditioning caused a significant upregulation of Timp2. Asterisks: regulation of the Ledgf and Timp2 was significantly different between light and hyperoxia conditioning (***P < 0.001 on a 2-tailed t-test).
Discussion
The present results contribute to the concept of the condition of the retina, the effect of environmental factors on the stability of photoreceptors. They confirm earlier reports that light exposure conditions the retina to be resistant to subsequent light stress, 46,30 and they show that the protective conditioning elicited by light exposure also protects against a different stress (hyperoxia). Conversely, they show that hyperoxic conditioning, sufficient to cause some damage to photoreceptors, does not elicit a protective response to subsequent challenge by either light or hyperoxia. In considering the implications of these findings for understanding of the ability of the retina to protect itself from environmental stress, it is useful to distinguish three relationships between environmental stress and photoreceptor stability. 
Damage and Protection Apparently Linked: Light and Hypoxic Stress
Several studies have described the effectiveness of exposure to bright light 5,29,31 and to hypoxia/ischemia 1317,3436,41 in making photoreceptors resistant to subsequent stress. Taken together, these observations could be viewed as evidence of a link between the damage caused by bright light and protective preconditioning, suggesting that protective mechanisms are upregulated as a consequence of photoreceptors being stressed or damaged. 
Damage without Protection: Hyperoxic Stress
The present observations, however, suggest that stress/damage to photoreceptors does not necessarily upregulate protective mechanisms. The levels of hyperoxia used for preconditioning in the present experiments induced a small but measurable rise in the rate of photoreceptor death in both C57BL/6J and Balb/cJ strains. We could not, however, demonstrate that hyperoxic stress reduced the impact of a subsequent hyperoxic challenge (in the C57BL/6J mouse) or light challenge (in the Balb/cJ mouse). Photoreceptors were damaged by hyperoxia, but protective mechanisms were not induced. 
Protection without Damage: Light Preconditioning in the C57BL/6J Strain
Conversely, in the light preconditioning experiment in the C57BL/6J strain (Figs. 3 45), significant exposure to light induced no increase in photoreceptor death and did not upregulate the stress-inducible proteins FGF-2 and GFAP, yet it did provide protection against subsequent light and hyperoxic stress. A recent study 42 in which prefeeding with saffron provided protection for rat photoreceptors against light damage could also be regarded as a preconditioning experiment. Saffron given as a dietary supplement provided protection without causing detectable stress to retinal structure or function. It thus appears that damage to photoreceptors is neither sufficient nor necessary for the upregulation of protective mechanisms. 
Hopeful Implications and Protective Mechanisms Explored
This analysis of the relationship between photoreceptor damage and protection has encouraging implications for therapy. If damage is not necessary for protection, then the potential exists to develop ways to elicit protection, without damage. The observation of Maccarone et al. 42 regarding dietary saffron is a step in this direction. They noted that another anti-oxidant (β-carotene), which is also a component of some foods, provides protection to photoreceptors but reduces their ability to respond to light. This observation draws attention to the fact that the mechanisms of protection are little known. The pioneering studies of Liu et al. 5,31 observed the upregulation of two trophic factors, FGF-2 and ciliary neurotrophic factor, by the preconditioning stress. In the present experiment, however, the upregulation of FGF-2 appeared unrelated to protection. For hypoxic stress, hypoxia-induced upregulation of erythropoietin may be important in the induction of protection. 34 Erythropoietin is hypoxia inducible and is unlikely to be upregulated by continuous light exposure, which raises oxygen levels in outer retina. 43  
The four genes whose expression in preconditioned retinas was examined by quantitative PCR were selected for their known neuroprotective actions and their upregulation by light. 26 As Huang et al. 26 noted, the 70-kDa heat shock protein Hsp70.1 is believed to play a critical role in cell survival in the face of stress. Exogenously delivered and elevated expression of HSP70 protects photoreceptors from light damage, 4446 and HSP70.1−/− mice are relatively vulnerable to photic injury 47 and hyperoxic stress. 45,46,48 FGF-13 is a member of the FGF (fibroblast growth factor) family, known to be protective to photoreceptors. 20,4951 Lens epithelium-derived growth factor (Ledgf) is a survival factor and transcriptional activator 52 that has been shown to protect photoreceptors in light-stressed and genetic rodent models of photoreceptor degeneration. 53,54 Tissue inhibitor of metalloproteinase 2 (Timp2) promotes cell growth and proliferation by activating intracellular cAMP signaling. 55 The present experiments confirmed that Hsp70.1, Fgf13, and Timp2 (but not Ledgf) are upregulated by light 26 and showed a major difference between light and hyperoxia regulation for one of the four genes, Timp2. The human ortholog of Timp2 mapped to the location of a known retinopathy locus listed on RetNet Web site (http://www.sph.uth.tmc.edu/Retnet/). 26 The exact role of Timp2 in protecting photoreceptors against stress after light preconditioning still must be defined. 
The Enigma of Hyperoxia
The present finding that hyperoxia damages photoreceptors but does not upregulate protective mechanisms was unexpected. It suggests that, in evolutionary terms, the retina has evolved light-inducible protective mechanisms because light absorption is (for obvious reasons) experienced by every animal with eyes, and its absorption is (as suggested by earlier reports 56,57 ) intrinsically stressful. The hypoxia-inducible protective mechanisms observed in the retina (above) may be shared with the other tissues of the body, in which ischemic induction of protective mechanisms was first demonstrated. Hyperoxia, arguably, is less common; it is perhaps for this reason that hyperoxia-inducible protective mechanisms have not evolved. 
Footnotes
 Supported by the Australian Research Council through the ARC Centre of Excellence in Vision Science (Grant CE0561903) and by the Sir Zelman Cowen Universities Fund.
Footnotes
 Disclosure: Y. Zhu, None; K. Valter, None; J. Stone, None
References
Penn JS Williams TP . Photostasis: regulation of daily photon-catch by rat retinas in response to various cyclic illuminances. Exp Eye Res. 1986;43:915–928. [CrossRef] [PubMed]
Penn JS . Early studies of the photostastis phenomenon. In: Williams TP Thistle AB eds. Photostasis and Related Phenomena. New York: Plenum Press; 1998:1–16.
Fulton AB Hansen RM Dodge J Williams TP . Photoreceptor development and photostasis. In: Williams TP Thistle AB eds. Photostasis and Related Phenomena. New York: Plenum Press; 1998:189–198.
Williams TP . Light history and photostasis. In: Williams TP Thistle AB eds. Photostasis and Related Phenomena. New York: Plenum Press; 1998:17–32.
Liu C Peng M Laties AM Wen R . Preconditioning with bright light evokes a protective response against light damage in the rat retina. J Neurosci. 1998;18:1337–1344. [PubMed]
Nir I Harrison JM Liu C Wen R . Extended photoreceptor viability by light stress in the RCS rats but not in the opsin P23H mutant rats. Invest Ophthalmol Vis Sci. 2001;42:842–849. [PubMed]
Kuwabara T Funahashi M . Light effect on the synaptic organ of the rat. Invest Ophthalmol. 1976;15:407–411. [PubMed]
Adly MA Spiwoks-Becker I Vollrath L . Ultrastructural changes of photoreceptor synaptic ribbons in relation to time of day and illumination. Invest Ophthalmol Vis Sci. 1999;40:2165–2172. [PubMed]
Spiwoks-Becker I Glas M Lasarzik I Vollrath L . Mouse photoreceptor synaptic ribbons lose and regain material in response to illumination changes. Eur J Neurosci. 2004;19:1559–1571. [CrossRef] [PubMed]
Gargini C Belfiore M Bisti S Cervetto L Valter K Stone J . The impact of bFGF on photoreceptor function and morphology. Invest Ophthalmol Vis Sci. 1999;40:2088–2099. [PubMed]
Gargini C Bisti S Demontis GC Valter K Stone J Cervetto L . Electroretinogram changes associated with retinal upregulation of trophic factors: observations following optic nerve section. Neuroscience. 2004;126:775–783. [CrossRef] [PubMed]
Valter K Bisti S Gargini C . Time course of neurotrophic factor upregulation and retinal protection against light-induced damage after optic nerve section. Invest Ophthalmol Vis Sci. 2005;46:1748–1754. [CrossRef] [PubMed]
Toprak A Ozbilgin K Toprak V Tuglu I Guler C . A histological analysis of the protective effect of ischemic preconditioning in the rat retina. Curr Eye Res. 2002;24:234–239. [CrossRef] [PubMed]
Zhang C Rosenbaum DM Shaikh AR . Ischemic preconditioning attenuates apoptotic cell death in the rat retina. Invest Ophthalmol Vis Sci. 2002;43:3059–3066. [PubMed]
Li Y Roth S Laser M Ma JX Crosson CE . Retinal preconditioning and the induction of heat-shock protein 27. Invest Ophthalmol Vis Sci. 2003;44:1299–1304. [CrossRef] [PubMed]
Grimm C Wenzel A Groszer M . HIF-1-induced erythropoietin in the hypoxic retina protects against light-induced retinal degeneration. Nat Med. 2002;8:718–724. [CrossRef] [PubMed]
Thiersch M Raffelsberger W Frigg R . Analysis of the retinal gene expression profile after hypoxic preconditioning identifies candidate genes for neuroprotection. BMC Genomics. 2008;9:73. [CrossRef] [PubMed]
Stone J Maslim J Valter-Kocsi K . Mechanisms of photoreceptor death and survival in mammalian retina. Prog Ret Eye Res. 1999;18:689–735. [CrossRef]
Noell WK . Visual cell effects of high oxygen pressures. Am Physiol Soc Fed Proc. 1955;14:107–108.
Yamada H Yamada E Ando A . Fibroblast growth factor-2 decreases hyperoxia-induced photoreceptor cell death in mice. Am J Pathol. 2001;159:1113–1120. [CrossRef] [PubMed]
Geller S Krowka R Valter K Stone J . Toxicity of hyperoxia to the retina: evidence from the mouse. Adv Exp Med Biol. 2006;572:425–437. [PubMed]
Natoli R Provis J Valter K Stone J . Expression and role of the early-response gene Oxr1 in the hyperoxia-challenged mouse retina. Invest Ophthalmol Vis Sci. 2008;49:4561–4567. [CrossRef] [PubMed]
Walsh N Bravo-Nuevo A Geller S Stone J . Resistance of photoreceptors in the C57BL/6-c2J, C57BL/6J, and BALB/cJ mouse strains to oxygen stress: evidence of an oxygen phenotype. Curr Eye Res. 2004;29:441–447. [CrossRef] [PubMed]
Gavrieli Y Sherman Y Ben-Sasson SA . Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation. J Cell Biol. 1992;119:493–501. [CrossRef] [PubMed]
Maslim J Valter K Egensperger R Hollander H Stone J . Tissue oxygen during a critical developmental period controls the death and survival of photoreceptors. Invest Ophthalmol Vis Sci. 1997;38:1667–1677. [PubMed]
Huang H Frank MB Dozmorov I . Identification of mouse retinal genes differentially regulated by dim and bright cyclic light rearing. Exp Eye Res. 2005;80:727–739. [CrossRef] [PubMed]
Pfaffl MW . A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res. 2001;29:e45. [CrossRef] [PubMed]
LaVail MM Gorrin GM Repaci MA . Strain differences in sensitivity to light-induced photoreceptor degeneration in albino mice. Curr Eye Res. 1987;6:825–834. [CrossRef] [PubMed]
Penn J Anderson R . Effects of light history on the rat retina. Prog Ret Res. 1991;11:75–98. [CrossRef]
Penn JS Tolman BL Thum LA Koutz CA . Effect of light history on the rat retina: time course of morphological adaptation and readaptation. Neurochem Res. 1992;17:91–99. [CrossRef] [PubMed]
Liu C Peng M Wen R . Pre-exposure to constant light protects photoreceptor from subsequent light damage in albino rats. Invest Ophthalmol Vis Sci. 1997;38:S718.
Kaldi I Martin RE Huang H Brush RS Morrison KA Anderson RE . Bright cyclic rearing protects albino mouse retina against acute light-induced apoptosis. Mol Vis. 2003;9:337–344. [PubMed]
Grimm C Wenzel A Acar N Keller S Seeliger M Gassmann M . Hypoxic preconditioning and erythropoietin protect retinal neurons from degeneration. Adv Exp Med Biol. 2006;588:119–131. [PubMed]
Grimm C Hermann DM Bogdanova A . Neuroprotection by hypoxic preconditioning: HIF-1 and erythropoietin protect from retinal degeneration. Semin Cell Dev Biol. 2005;16:531–538. [CrossRef] [PubMed]
Ozbay D Ozden S Muftuoglu S . Protective effect of ischemic preconditioning on retinal ischemia-reperfusion injury in rats. Can J Ophthalmol. 2004;39:727–732. [CrossRef] [PubMed]
Kamphuis W Dijk F Bergen AA . Ischemic preconditioning alters the pattern of gene expression changes in response to full retinal ischemia. Mol Vis. 2007;13:1892–1901. [PubMed]
Walsh N Valter K Stone J . Cellular and subcellular patterns of expression of bFGF and CNTF in the normal and light stressed adult rat retina. Exp Eye Res. 2001;72:495–501. [CrossRef] [PubMed]
Bowers F Valter K Chan S Walsh N Maslim J Stone J . Effects of oxygen and bFGF on the vulnerability of photoreceptors to light damage. Invest Ophthalmol Vis Sci. 2001;42:804–815. [PubMed]
Esposito LA Melov S Panov A Cottrell BA Wallace DC . Mitochondrial disease in mouse results in increased oxidative stress. Proc Natl Acad Sci U S A. 1999;96:4820–4825. [CrossRef] [PubMed]
Wallace DC . Mouse models for mitochondrial disease. Am J Med Genet. 2001;106:71–93. [CrossRef] [PubMed]
Casson RJ Wood JP Melena J Chidlow G Osborne NN . The effect of ischemic preconditioning on light-induced photoreceptor injury. Invest Ophthalmol Vis Sci. 2003;44:1348–1354. [CrossRef] [PubMed]
Maccarone R Di Marco S Bisti S . Saffron supplement maintains morphology and function after exposure to damaging light in mammalian retina. Invest Ophthalmol Vis Sci. 2008;49:1254–1261. [CrossRef] [PubMed]
Braun R Linsenmeier R Goldstick T . Light-induced changes in retinal oxygen consumption. Invest Ophthalmol Vis Sci. 1996;37:473–475. [PubMed]
Barbe MF Tytell M Gower DJ Welch WJ . Hyperthermia protects against light damage in the rat retina. Science. 1988;241:1817–1820. [CrossRef] [PubMed]
Yu Q Kent CR Tytell M . Retinal uptake of intravitreally injected Hsc/Hsp70 and its effect on susceptibility to light damage. Mol Vis. 2001;7:48–56. [PubMed]
Yu YS Heo JH Hwang SW . Effect of the absence of heat shock protein 70.1 (hsp70.1) on retinal photoreceptors in normal and rd mice. Korean J Ophthalmol. 2001;15:67–73. [CrossRef] [PubMed]
Kim JH Yu YS Chung H Heo JW Seo JS . Effect of the absence of heat shock protein 70.1 (hsp70.1) on retinal photic injury. Korean J Ophthalmol. 2003;17:7–13. [CrossRef] [PubMed]
Choi MY Yu YS Kim SK Kim YJ Seo JS . The effect of oxygen on retinal degeneration in wild-type and hsp70.1 knockout neonatal retinal degeneration mice. Korean J Ophthalmol. 2001;15:1–7. [CrossRef] [PubMed]
Gao H Hollyfield JG . Basic fibroblast growth factor: increased gene expression in inherited and light-induced photoreceptor degeneration. Exp Eye Res. 1996;62:181–189. [CrossRef] [PubMed]
Yao DL Masonic K Petullo D . Pretreatment with intravenous FGF-13 reduces infarct volume and ameliorates neurological deficits following focal cerebral ischemia in rats. Brain Res. 1999;818:140–146. [CrossRef] [PubMed]
Faktorovich EG Steinberg RH Yasumura D Matthes MT LaVail MM . Photoreceptor degeneration in inherited retinal dystrophy delayed by basic fibroblast growth factor. Nature. 1990;347:83–86. [CrossRef] [PubMed]
Shinohara T Singh DP Fatma N . LEDGF, a survival factor, activates stress-related genes. Prog Retin Eye Res. 2002;21:341–358. [CrossRef] [PubMed]
Machida S Chaudhry P Shinohara T . Lens epithelium-derived growth factor promotes photoreceptor survival in light-damaged and RCS rats. Invest Ophthalmol Vis Sci. 2001;42:1087–1095. [PubMed]
Ahuja P Caffe AR Holmqvist I . Lens epithelium-derived growth factor (LEDGF) delays photoreceptor degeneration in explants of rd/rd mouse retina. Neuroreport. 2001;12:2951–2955. [CrossRef] [PubMed]
Corcoran ML Stetler-Stevenson WG . Tissue inhibitor of metalloproteinase-2 stimulates fibroblast proliferation via a cAMP-dependent mechanism. J Biol Chem. 1995;270:13453–13459. [CrossRef] [PubMed]
Noell WK Walker VS Kang BS Berman S . Retinal damage by light in rats. Invest Ophthalmol. 1966;5:450–473. [PubMed]
Chrysostomou V Stone J Stowe S Barnett NL Valter K . The status of cones in the rhodopsin mutant P23H-3 retina: light-regulated damage and repair in parallel with rods. Invest Ophthalmol Vis Sci. 2008;49:1116–1125. [CrossRef] [PubMed]
Figure 1.
 
Experimental design. Two strains of mice were studied: C57BL/6J because their photoreceptors are hyperoxia vulnerable and BALB/cJ because their photoreceptors are light vulnerable. In each strain, three groups were studied: control, light-preconditioned, and hyperoxia-preconditioned. For the C57BL/6J strain, the test stress was hyperoxia; for the BALB/cJ strain, the test stress was bright light.
Figure 1.
 
Experimental design. Two strains of mice were studied: C57BL/6J because their photoreceptors are hyperoxia vulnerable and BALB/cJ because their photoreceptors are light vulnerable. In each strain, three groups were studied: control, light-preconditioned, and hyperoxia-preconditioned. For the C57BL/6J strain, the test stress was hyperoxia; for the BALB/cJ strain, the test stress was bright light.
Figure 2.
 
Sections of mouse retina labeled for dying cell (red) with the TUNEL technique and for surviving cells (blue) with bisbenzimide. (A, D) Control retinas of C57BL/6J and Balb/cJ strains showed few TUNEL-positive cells. (B, E) Two weeks' exposure to 75% hyperoxia caused a large increase in the number of TUNEL-positive cells in the ONL of the C57BL/6J mouse and a lesser increase in the Balb/cJ mouse. The ONL was thinned in the C57BL/6J mouse. (C, F) One week's exposure to 1000 lux light caused no measurable rise in the number of TUNEL-positive cells in the C57BL/6J mouse but increased TUNEL-positive cell numbers in the ONL of the Balb/cJ mouse. The ONL was thinned by light exposure in the Balb/cJ mouse. Scale bar, 20 μm (A–F).
Figure 2.
 
Sections of mouse retina labeled for dying cell (red) with the TUNEL technique and for surviving cells (blue) with bisbenzimide. (A, D) Control retinas of C57BL/6J and Balb/cJ strains showed few TUNEL-positive cells. (B, E) Two weeks' exposure to 75% hyperoxia caused a large increase in the number of TUNEL-positive cells in the ONL of the C57BL/6J mouse and a lesser increase in the Balb/cJ mouse. The ONL was thinned in the C57BL/6J mouse. (C, F) One week's exposure to 1000 lux light caused no measurable rise in the number of TUNEL-positive cells in the C57BL/6J mouse but increased TUNEL-positive cell numbers in the ONL of the Balb/cJ mouse. The ONL was thinned by light exposure in the Balb/cJ mouse. Scale bar, 20 μm (A–F).
Figure 3.
 
Effects of light preconditioning on light-induced cell death in Balb/cJ retina. Four groups of animals were analyzed (each group contained five animals). (A–D) TUNEL labeling (red) shows dying cells in the sensitive superior region of the retina of Balb/cJ mice after different treatment. The nuclei of retinal cells are labeled blue with bisbenzimide. The number of TUNEL-positive profiles was not raised by 2 days' exposure to 300 lux (A, B) but was markedly raised by 1 week's exposure to 1000 lux (A, C). The light-induced rise was reduced by light preconditioning (D). (E) Mean numbers of TUNEL-positive profiles/section in the ONL of control and treated Balb/cJ retinas. Each symbol represents the mean for one animal; the horizontal line marks the mean of each group; the error bars represent ± SEM. Preconditioning (300 lux for 2 days) did not increase TUNEL labeling; the test stress (1000 lux for 1 week) produced a very significant increase. This increase was reduced by preconditioning, and the reduction was statistically significant (P = 0.001). (F) ONL thickness, estimated by the ONL/INL ratio, plotted as a function of distance from the optic nerve head (ONH) along the vertical meridian of the eye. The mean value for each group is plotted, with error bars representing the SEM of measures at each point. After 1-week exposure of damaging light, the ONL thickness was reduced in the unconditioned retina, most markedly just superior to the optic disc. The reduction was decreased by preconditioning. Asterisks: points at which the ONL thickness in the conditioned light-damaged group (300 lux for 2 days, then 1000 lux for 1 week) was significantly (P < 0.05 on a 2-tailed t-test) greater than in the light-damaged group (1000 lux for 1 week). That is, these are the points at which the rescue of the ONL reached statistical significance. Scale bar, 20 μm (A–D).
Figure 3.
 
Effects of light preconditioning on light-induced cell death in Balb/cJ retina. Four groups of animals were analyzed (each group contained five animals). (A–D) TUNEL labeling (red) shows dying cells in the sensitive superior region of the retina of Balb/cJ mice after different treatment. The nuclei of retinal cells are labeled blue with bisbenzimide. The number of TUNEL-positive profiles was not raised by 2 days' exposure to 300 lux (A, B) but was markedly raised by 1 week's exposure to 1000 lux (A, C). The light-induced rise was reduced by light preconditioning (D). (E) Mean numbers of TUNEL-positive profiles/section in the ONL of control and treated Balb/cJ retinas. Each symbol represents the mean for one animal; the horizontal line marks the mean of each group; the error bars represent ± SEM. Preconditioning (300 lux for 2 days) did not increase TUNEL labeling; the test stress (1000 lux for 1 week) produced a very significant increase. This increase was reduced by preconditioning, and the reduction was statistically significant (P = 0.001). (F) ONL thickness, estimated by the ONL/INL ratio, plotted as a function of distance from the optic nerve head (ONH) along the vertical meridian of the eye. The mean value for each group is plotted, with error bars representing the SEM of measures at each point. After 1-week exposure of damaging light, the ONL thickness was reduced in the unconditioned retina, most markedly just superior to the optic disc. The reduction was decreased by preconditioning. Asterisks: points at which the ONL thickness in the conditioned light-damaged group (300 lux for 2 days, then 1000 lux for 1 week) was significantly (P < 0.05 on a 2-tailed t-test) greater than in the light-damaged group (1000 lux for 1 week). That is, these are the points at which the rescue of the ONL reached statistical significance. Scale bar, 20 μm (A–D).
Figure 4.
 
Effects of light preconditioning on hyperoxia-induced cell death in C57BL/6J retina. Four groups of animals were analyzed (each group contained six to nine animals). (A–D) Representative pictures of TUNEL staining in the inferior retina of C57BL/6J mice after different treatment. The number of TUNEL-positive profiles was not raised by 1 week's exposure to 1000 lux (A, B) and was markedly raised by 2 weeks' exposure to 75% O2 (A, C). The hyperoxia-induced increase was reduced by previous exposure to light (D). (E) Mean number of TUNEL-positive profiles/section in the ONL of control and treated C57BL/6J retinas. Each symbol represents the mean for one animal, the horizontal line marks the mean of each group, and the error bars represent ± SEM. Preconditioning (1000 lux for 1 week) did not increase TUNEL labeling; the test stress (75% oxygen for 2 weeks) produced a very significant increase. This increase was reduced by preconditioning, and the reduction was statistically significant (P < 0.01). (F) ONL thickness, estimated by the ONL/INL ratio, plotted as a function of distance from the optic nerve head (ONH) along the vertical meridian of the eye. The mean value for each group is plotted, with error bars representing the SEM of measures at each point. After 2 weeks' exposure to 75% oxygen, the ONL thickness was reduced in the unconditioned retina, most markedly just inferior to the optic disc. Asterisks: points at which the thickness of the ONL in the conditioned hyperoxia-damaged group (1000 lux for 1 week, then 75% O2 for 2 weeks) was significantly (P < 0.05 on a 2-tailed t-test) greater than in the unconditioned hyperoxia-damaged group (75% O2 for 2 weeks). That is, these are the points at which the rescue of the ONL reached statistical significance. Scale bar, 20 μm (A–D).
Figure 4.
 
Effects of light preconditioning on hyperoxia-induced cell death in C57BL/6J retina. Four groups of animals were analyzed (each group contained six to nine animals). (A–D) Representative pictures of TUNEL staining in the inferior retina of C57BL/6J mice after different treatment. The number of TUNEL-positive profiles was not raised by 1 week's exposure to 1000 lux (A, B) and was markedly raised by 2 weeks' exposure to 75% O2 (A, C). The hyperoxia-induced increase was reduced by previous exposure to light (D). (E) Mean number of TUNEL-positive profiles/section in the ONL of control and treated C57BL/6J retinas. Each symbol represents the mean for one animal, the horizontal line marks the mean of each group, and the error bars represent ± SEM. Preconditioning (1000 lux for 1 week) did not increase TUNEL labeling; the test stress (75% oxygen for 2 weeks) produced a very significant increase. This increase was reduced by preconditioning, and the reduction was statistically significant (P < 0.01). (F) ONL thickness, estimated by the ONL/INL ratio, plotted as a function of distance from the optic nerve head (ONH) along the vertical meridian of the eye. The mean value for each group is plotted, with error bars representing the SEM of measures at each point. After 2 weeks' exposure to 75% oxygen, the ONL thickness was reduced in the unconditioned retina, most markedly just inferior to the optic disc. Asterisks: points at which the thickness of the ONL in the conditioned hyperoxia-damaged group (1000 lux for 1 week, then 75% O2 for 2 weeks) was significantly (P < 0.05 on a 2-tailed t-test) greater than in the unconditioned hyperoxia-damaged group (75% O2 for 2 weeks). That is, these are the points at which the rescue of the ONL reached statistical significance. Scale bar, 20 μm (A–D).
Figure 5.
 
Effects of light preconditioning on the expression of stress-related proteins (A–F) and on OS length (K–N) in the C57BL/6J retina. (A–F) GFAP (red) is confined to astrocytes at the inner surface of the control retina (A). After 2 weeks' hyperoxia, its expression was upregulated in Müller cells in the inferior retina (D). This reduction was decreased by light preconditioning (F). FGF-2 (green) was not prominent in the ONL in control retinas but was upregulated after 2 weeks' hyperoxia (in inferior retina, D). This upregulation was reduced by light preconditioning (F). (G–J) Acrolein labeling in the vulnerable inferior region was upregulated after 2 weeks' hyperoxia (I). The upregulation was reduced by light preconditioning (J). (K–N) MnSOD labeling in the vulnerable region of inferior retina. Labeling of IS was upregulated after 2 weeks' hyperoxia; the upregulation was reduced by light preconditioning. (O–R) Labeling for rhodopsin (red) label and L/M opsin (green) in the vulnerable region of the inferior retina indicates that the lengths of rod and cone OS were reduced by 2 weeks' hyperoxia; the reduction was mitigated by light preconditioning. GCL, ganglion cell layer; RPE, retinal pigment epithelium. Scale bars: 20 μm (A–N, O–R).
Figure 5.
 
Effects of light preconditioning on the expression of stress-related proteins (A–F) and on OS length (K–N) in the C57BL/6J retina. (A–F) GFAP (red) is confined to astrocytes at the inner surface of the control retina (A). After 2 weeks' hyperoxia, its expression was upregulated in Müller cells in the inferior retina (D). This reduction was decreased by light preconditioning (F). FGF-2 (green) was not prominent in the ONL in control retinas but was upregulated after 2 weeks' hyperoxia (in inferior retina, D). This upregulation was reduced by light preconditioning (F). (G–J) Acrolein labeling in the vulnerable inferior region was upregulated after 2 weeks' hyperoxia (I). The upregulation was reduced by light preconditioning (J). (K–N) MnSOD labeling in the vulnerable region of inferior retina. Labeling of IS was upregulated after 2 weeks' hyperoxia; the upregulation was reduced by light preconditioning. (O–R) Labeling for rhodopsin (red) label and L/M opsin (green) in the vulnerable region of the inferior retina indicates that the lengths of rod and cone OS were reduced by 2 weeks' hyperoxia; the reduction was mitigated by light preconditioning. GCL, ganglion cell layer; RPE, retinal pigment epithelium. Scale bars: 20 μm (A–N, O–R).
Figure 6.
 
Effects of hyperoxic preconditioning on hyperoxia-induced photoreceptor death in the C57BL/6J mouse (A) and on light- induced photoreceptor death in the Balb/cJ mouse (B). Four groups of animals were examined for each series (each group contained five animals). Mean values and SEMs are shown for each group. Asterisks: the increased frequency of TUNEL-positive cells caused by hyperoxic preconditioning was significantly (*P < 0.001) greater than in control. Hyperoxic preconditioning did not cause a reduction in the frequency of TUNEL labeling induced by either hyperoxia (A) or light (B) test stress.
Figure 6.
 
Effects of hyperoxic preconditioning on hyperoxia-induced photoreceptor death in the C57BL/6J mouse (A) and on light- induced photoreceptor death in the Balb/cJ mouse (B). Four groups of animals were examined for each series (each group contained five animals). Mean values and SEMs are shown for each group. Asterisks: the increased frequency of TUNEL-positive cells caused by hyperoxic preconditioning was significantly (*P < 0.001) greater than in control. Hyperoxic preconditioning did not cause a reduction in the frequency of TUNEL labeling induced by either hyperoxia (A) or light (B) test stress.
Figure 7.
 
Effects of hyperoxia preconditioning on retinal expression of the stress-related proteins GFAP (red) and FGF-2 (green). (A–F) 75% hyperoxia for 2 weeks in the C57BL/6J retina upregulated the expression of GFAP in Müller cells (red) and of FGF-2 (green) in the ONL inferior retina (F). The upregulation of FGF-2, but not of GFAP, appeared to be mitigated by previous exposure to hyperoxia. (G–J) Light exposure of the Balb/cJ retina caused a limited upregulation of GFAP in Müller cells, in both the superior and the inferior retina (I) that was more prominent after hyperoxia preconditioning (J). FGF-2 expression in the ONL was also most prominent after the retina was exposed to both hyperoxia preconditioning and light stress (J). Scale bar, 20 μm (A–J).
Figure 7.
 
Effects of hyperoxia preconditioning on retinal expression of the stress-related proteins GFAP (red) and FGF-2 (green). (A–F) 75% hyperoxia for 2 weeks in the C57BL/6J retina upregulated the expression of GFAP in Müller cells (red) and of FGF-2 (green) in the ONL inferior retina (F). The upregulation of FGF-2, but not of GFAP, appeared to be mitigated by previous exposure to hyperoxia. (G–J) Light exposure of the Balb/cJ retina caused a limited upregulation of GFAP in Müller cells, in both the superior and the inferior retina (I) that was more prominent after hyperoxia preconditioning (J). FGF-2 expression in the ONL was also most prominent after the retina was exposed to both hyperoxia preconditioning and light stress (J). Scale bar, 20 μm (A–J).
Figure 8.
 
Quantitative real-time PCR analysis of potential neuroprotective genes in C57BL/6J retina after light and hyperoxia preconditioning. The expression level of genes in preconditioned groups was compared to that in the control group, and a value was generated for the fold-change in expression. A fold-change of 1 (dashed line) indicates no change in expression of the gene, a fold-change >1 indicates upregulation in the experimental group, and a fold-change <1 indicates downregulation. Four biological replicates were used in each precondition, with the error bar indicating the SEM. Hyperoxia caused only limited regulation of the four genes tested (Rho was selected as a retinally expressed gene whose expression is not known to be regulated by hyperoxia or light). Light preconditioning caused a significant upregulation of Timp2. Asterisks: regulation of the Ledgf and Timp2 was significantly different between light and hyperoxia conditioning (***P < 0.001 on a 2-tailed t-test).
Figure 8.
 
Quantitative real-time PCR analysis of potential neuroprotective genes in C57BL/6J retina after light and hyperoxia preconditioning. The expression level of genes in preconditioned groups was compared to that in the control group, and a value was generated for the fold-change in expression. A fold-change of 1 (dashed line) indicates no change in expression of the gene, a fold-change >1 indicates upregulation in the experimental group, and a fold-change <1 indicates downregulation. Four biological replicates were used in each precondition, with the error bar indicating the SEM. Hyperoxia caused only limited regulation of the four genes tested (Rho was selected as a retinally expressed gene whose expression is not known to be regulated by hyperoxia or light). Light preconditioning caused a significant upregulation of Timp2. Asterisks: regulation of the Ledgf and Timp2 was significantly different between light and hyperoxia conditioning (***P < 0.001 on a 2-tailed t-test).
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×