Molecular Vision 2005; 11:309-320 <http://www.molvis.org/molvis/v11/a36/> Received 31 August 2004 | Accepted 4 April 2005 | Published 28 April 2005

Download Reprint

Regulation of Egr-1, VIP, and Shh mRNA and Egr-1 protein in the mouse retina by light and image quality

Christine Brand,¹ Eva Burkhardt,¹ Frank Schaeffel,¹ Jeong Won Choi,² Marita Pauline Feldkaemper¹
 
 

¹Section for Neurobiology of the Eye, University Eye Hospital Tuebingen, Tuebingen, Germany; ²Department of Biology, City College, City University of New York, New York, NY

Correspondence to: Dr. Marita Feldkaemper, Section for Neurobiology of the Eye, University Eye Hospital, University of Tuebingen, Calwerstrasse 7/1, 72076 Tuebingen, Germany; Phone: (+49) 7071-29-87424; FAX: (+49) 7071-29-5196; e-mail: marita.feldkaemper@uni-tuebingen.de

Abstract

Purpose: To analyze mRNA expression changes of Egr-1, VIP, and Shh under different light and treatment conditions in mice. The mRNA expression levels of the three genes and additionally the Egr-1 protein expression were compared in form deprived eyes and eyes with normal vision. Moreover, the influence of dark to light and light to dark transitions and of changes in retinal illumination on mRNA levels was investigated.

Methods: Form deprivation of mice was induced by fitting frosted diffusers over one eye and an attentuation matched neutral density (ND) filter over the other eye. To measure the effects of retinal illumination changes on mRNA expression, animals were bilaterally fitted with different ND filters. Semiquantitative real-time RT-PCR was used to measure the mRNA levels and immunohistochemistry was applied to localize and detect Egr-1 protein.

Results: The expression levels of both Egr-1 mRNA and protein were reduced in form deprived eyes compared to their fellow eyes after 30 min and 1 h, respectively. Egr-1 mRNA was strikingly upregulated both after dark to light and light to dark transitions, whereas minor changes in retinal illumination by covering the eyes with neutral density filters did not alter Egr-1 mRNA expression. In mice, the mRNA levels of VIP and Shh were not affected by form deprivation, but they were found to be regulated depending on the time of day.

Conclusions: Both Egr-1 mRNA and protein expression levels were strongly regulated by light, especially by transitions between light and darkness. Image contrast may exert an additional influence on mRNA and protein expression of Egr-1, particularly in the cells in the ganglion cell layer and in bipolar cells.

Introduction

In the chick, myopia and hyperopia can be artificially induced by treating the birds with negative and positive lenses, respectively. Moreover, treating chicks with diffusers that deprive the retina from high contrast and high spatial frequencies also leads to myopia. A local mechanism within the retina is involved in eye growth regulation since neither accommodation nor contributions from the brain are necessary [1]. Among the biochemical changes induced by lens and diffuser treatment in chicks and tree shrews were alterations in the expression levels of the transcription factor Egr-1 (also called ZENK), VIP (Vasoactive Intestinal Polypeptide), and Shh (Sonic hedgehog). Recently, a mouse model for myopia was established [2,3]. The major advantages of the mouse are the completely sequenced genome and the presence of numerous knockout models. Since the biochemical pathways underlying eye growth regulation involve Egr-1, VIP, and Shh in the chick, in this report we studied their visual regulation in the mouse.

Early growth response protein 1 (Egr-1) [4], also known as NGFI-A, zif268, tis8, cef5, Krox24, and the acronym "ZENK", was first identified as an immediate early gene responsive to growth factors and various differentiation signals and later confirmed as a transcriptional regulatory protein. Egr-1 is induced in the absence of de novo protein synthesis by mitogens, developmental or differentiation cues, tissue or radiation injury, or signals that cause neuronal excitation [5]. Located in the nucleus [6-8], Egr-1 is a zinc-finger protein with three tandemly repeated Cys₂His₂ zinc-finger motifs and has, according to its role as a transcription factor, numerous target genes [9], among which are PDGF-A [10] and B [11], bFGF [12] and TGF-β1 [11,13]. A potential connection between Egr-1 and myopia was first described in the chick, where it was found that the expression of ZENK correlates with the sign of defocus imposed by lenses in a subset of amacrine cells (AC), specifically the glucagon AC. Moreover, the number of ZENK-expressing bipolar cells (BC) and AC (all kinds, including the glucagon AC) varied with time of day [14]. Only recently, focus sensitive immunoreactivity for Egr-1 was shown in a subpopulation of GABAergic amacrine cells (GAD65-immunoreactive cells) in the macaque retina [15].

Vasoactive intestinal polypeptide (VIP) was first isolated from porcine intestinal wall as a 28 amino acid peptide related to glucagon and secretin [16]. It was found that VIP stimulates the retinal adenylate cyclase in rabbit [17] and induces formation of cyclic AMP in vitro [18] in the rabbit, rat and calf. VIP containing neurons (amacrine cells) could be found in the chick retina [19] and in the retinas of different species like mammals, frogs, and fish [20]. A possible link between VIP expression and regulation of ocular growth was suggested by the observation that the levels of VIP are increased in deprived eyes of primates when compared to open eyes [21]. In the chick, VIP may be involved both in the normal development of ocular refraction and in the development of form deprivation myopia [22].

Finally, sonic hedgehog (Shh) is a secreted protein implicated in the regulation of CNS polarity and related to the Drosophila segment polarity gene hedgehog (hh) [23]. It is expressed in the murine retina from P7 onwards until adulthood. Shh is localized both in the ganglion cell layer and in a subset of cells in the inner half of the inner nuclear layer, possibly amacrine cells in mouse [24]. A recent study proposes that the Shh pathway may act as a regulator of both prenatal and postnatal retinal growth in mice [25]. In chicks, both mRNA and protein expression levels of Shh were found to increase under conditions that induce myopia development. This finding suggests a potential involvement of Shh in the retinal control of postnatal eye growth [26,27].

The purpose of the present study was to gain information about the roles of these signaling molecules in eye growth regulation in mice. Both mRNA and protein expression of Egr-1, and mRNA levels of VIP and Shh were measured in form deprived mice and mice with normal visual experience. Moreover, the influence of retinal image illumination and light-dark transitions was studied.

Methods

Animals

Black wildtype C57BL/6 mice were raised under a 12 h light: 12 h dark cycle with light onset at 8:00 AM and were studied at postnatal age P29 to P33. The animals were reared in the local animal facility with free access to water and food. The experimental treatment was in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. The experiments were approved by the University Commission for Animal Welfare (reference AK 3/02).

Light conditions during experimental procedures

During the experiments (1) through (4), animals were kept in translucent plastic boxes with wire tops and without additional visual features under a fume hood illuminated by cool white light (Lumilux 30 W/840; Osram, Munich, Germany). Overall illumination was about 120 lux, which is about 2 log units below the illumination levels used to induce retinal degeneration in mice [28].

Treatment 1: Gene and protein expression in the retinas of control animals over time

To follow the changes in mRNA levels over the day, untreated mice were kept on a regular 12 h light/dark cycle and sampled at different time points within this cycle: before light onset in the morning, 15, 30, 120, 360, or 720 min after light onset and 30 min after light offset in the evening. Each group consisted of 5 age matched animals (mainly litter mates).

For protein expression analysis, mice were also kept on a regular 12 h light/dark cycle and sampled at different time points: before light onset in the morning, 1, 4, or 6.5 h after light onset and 1 h after light offset in the evening. For histological studies, all groups consisted of 6 age matched animals.

Different light exposure periods for mRNA and protein analyses were chosen since changes in mRNA levels can be detected more rapidly than changes in protein levels.

Treatment 2: Analysis of retinal Egr-1 protein expression in animals after defined light/dark conditions

To investigate the influence of light on Egr-1 protein expression, two groups of animals were kept under special light/dark conditions consisting of four alternating light and dark periods, more precisely: light-dark-light-dark. The length of a single period was 90 min and the first light period began right after the regular 12 h dark cycle. The animals of one group were sacrificed directly after the last dark period, whereas the animals of the other group were exposed to light for 1 h after the last dark period before tissue preparation.

To investigate the influence of the time of day and light on Egr-1 protein expression, another group of animals was kept in darkness until 1:30 p.m. and was then exposed to light for 1 h before tissue preparation.

Treatment 3: Effects of defined light attenuation by neutral density (ND) filters on gene expression in the retina

Experiments were conducted to investigate the influence of retinal image illumination on mRNA expression: animals (n=5 for each experiment) were fitted with ND filters of different light attenuation (Kodak Wratten Gelatin Filter No. 96, Rochester, NY). ND filters with 0.1, 0.3 and 0.5 log units attenuation (corresponding to 80, 50 and 32% transmission, respectively) were bilaterally attached to the periorbital fur by superglue (UHU instant glue "Sekundenkleber", Gel; Buehl, Germany) in the evening. An additional group was bilaterally fitted with completely black diffusers. After the attachment of the ND filters and diffusers animals were immediately put into darkness. All animals were exposed to 30 min of light in the morning of the following day before retinal gene expression was measured.

Treatment 4: Effects of image degradation on the expression of Egr-1, VIP, and Shh mRNA and Egr-1 protein

To investigate the influence of spatial information in the retinal image on mRNA expression, mice (n=5) were fitted with frosted diffusers over one eye and ND filters over the other (0.3 log units light attenuation) in the evening. They were kept in darkness overnight. We used heavily frosted diffusers, as described earlier [29]. These diffusers act as low pass filters on the spatial frequency spectrum, and reduce contrast over a wide range of spatial frequencies. Light attenuation through both diffusers and ND filters was matched (as measured by a light meter, Nuernberg, Germany) so that the illumination was balanced in both eyes. Starting at 8:00 AM, single animals were consecutively transferred into light and were exposed for either 15, 30, 120, or 360 min.

For Egr-1 protein expression analysis, mice (n=6) were also fitted with diffusers and ND filters and exposed to light for 60 min.

Treatment 5: Is the expression of Egr-1 controlled by spatial features in the retinal image?

An additional experiment was conducted to find out whether Egr-1 expression is controlled only by retinal illumination or also by spatial features. To this end, the eyes of six mice were covered with a diffuser on one side and an attenuation matched ND filter on the other side. Differences in Egr-1 expression can then not be attributed to differences in retinal illumination. The mice were exposed to light for 5.5 h before the ND filters and diffusers were attached so that the Egr-1 expression level was low at the beginning of the experiment. After attachment of the diffusers and ND filters, animals were kept in a rotating drum with a diameter of 60 cm and height 35 cm for 30 min. Illumination was about 400 lux, the stripe pattern had a spatial frequency of 0.1 cycles per degree, and the angular velocity was about 1 cyc/s.

Diffuser and ND filter design

Preliminary experiments were performed to control for slight differences in the design of diffusers and ND filters. The optomotor grating acuity of the C57BL/6 mice is limited to approximately 0.3 (cycles/degree) [30]. Previous experiments have shown that the image contrast is reduced to about 60-70% of the initial value at the relevant spatial frequencies [29]. Diffusers were hand made from transparencies that were frosted by emery paper. The hemispherical shaping of the diffuser was achieved by pushing a heated metal ball against the transparency. ND filters were glued to translucent supports since they were made from gelatin and could therefore not be shaped like the diffusers. Preliminary experiments with two kinds of diffusers ("bulged" diffusers made from one piece of frosted transparency and "plain" diffusers made from frosted transparency affixed to a black support) showed no differences in Egr-1 mRNA expression. The same result was found for ND filters glued to either translucent or black supports. It is likely that the photoreceptors compensate for image illumination differences by adaptation.

Tissue preparation, RNA isolation, and cDNA synthesis

All animals were sacrificed by an overdose of diethylether at the end of the respective light or dark period. Eyes were enucleated and immediately put into a petri dish filled with ice chilled Ringer solution until preparation. The eyes were consecutively put onto a filter paper, perforated by a canula and opened with scissors cutting around the iris. The lens was removed and the retina was extracted, snap frozen in liquid nitrogen and stored at -70 °C. Total RNA was isolated with a kit (RNeasy Mini Kit; Qiagen, Hilden, Germany) according to the manufacturer's instructions. The retinas were homogenized for 1 min with speed increasing from 11,000 to 20,000 rpm (Diax 900 Homogenizer; Heidolph, Kelheim, Germany). All RNA samples were treated with RNase free DNase I (Roche, Mannheim, Germany). The quality of RNA was checked by gel electrophoresis, concentration and purity were determined by spectrophotometry at 260 and 280 nm (average ratio 1.89±0.1). Of each sample, 0.25 μg total RNA were reverse transcribed (SuperScript^TMII RNase H^- reverse transcriptase; Invitrogen, Paisley, UK) using a mixture of 50 ng random hexamers and 500 ng oligo (dT)₁₅ primers in a total volume of 20 μl.

Histology

Eyes were enucleated and eyecups were prepared in Ringer solution by cutting around the iris, opening the eye and removing the lens. The applied fixation method is described elsewhere [31].

Antibodies and their working dilutions included anti-Egr-1, rabbit polyclonal antibody at 1:1250 (Egr-1 (588), sc-110, Lot number L239, Santa Cruz Biotechnology Inc., Santa Cruz, CA), anti-glutamine synthetase, mouse monoclonal antibody at 1:500 (MAB302, Chemicon, Temecula, CA), anti-protein kinase C, mouse monoclonal antibody at 1:400 (PKC (MC5), sc-80, Santa Cruz), and anti-recoverin polyclonal rabbit antibody at 1:500 (gift from Dr. K. W. Koch from the Universitiy of Oldenburg, Germany). Using mouse monoclonal antibodies for the labeling of mouse tissue might lead to a higher background, but a successful application of primary antibodies from the same species was previously shown [32]. In double labeling experiments, sections were incubated with a mixture of primary antibodies for glutamine synthetase and Egr-1, followed by a mixture of secondary antibodies. Cyanin 3 (Cy3-linked rabbit anti-mouse IgG, Amersham Biosciences, Freiburg, Germany) was used as fluorophore for both Egr-1 and recoverin, and Oregon green (goat anti-mouse IgG, Molecular Probes) for glutamine synthetase and protein kinase C labeling.

Figure 1A,B show Egr-1 immunoreactive cells in the mouse retina using diaminobenzidine (Figure 1A) and a fluorescent secondary antibody for labeling (Figure 1B), respectively. Cells in the ganglion cell layer (GCL), amacrine cells (AC), putative bipolar cells (BC) in the inner nuclear layer, and photoreceptor terminals were intensely labeled. Within the ganglion cell layer, it was not possible to differentiate between ganglion cells and displaced amacrine cells. To clarify the nature of the cells in the inner nuclear layer, retinal sections were labeled with antibodies for glutamine synthetase (GS), protein kinase C (PKC) and recoverin. Glutamine synthetase labeling was restricted to Müller cells (Figure 1C) [32] and the double labeling showed that Müller cells were not immunoreactive for Egr-1 (Figure 1D). Bipolar cells and some amacrine cells labeled for protein kinase C (Figure 1E). Recoverin labeled ON and OFF cone bipolar cells (Figure 1F).

For the analysis of Egr-1 protein expression, frozen eye cups vertically were cut until the optic nerve became visible. Four sections derived from the optic nerve region were analyzed per eye. The labeled cells were counted in four microscope fields per section under 400x magnification. The four microscope fields consisted of two central and two peripheral fields. The center (C) was located as the two fields adjacent to the optic disc, and the periphery (P) as the following microscope fields (Figure 2). The four fields almost covered a whole retinal section. Cell counts given in the figures correspond to average cell counts in the four microscope fields. The counting of the Egr-1 immunoreactive cells was performed in a masked fashion.

Semiquantitative real-time RT-PCR

The sequences of 18S rRNA (X00686), β-actin (M12481), Egr-1 (M20157), VIP (NM_011702) and Shh (NM_009170) were obtained from the National Center for Biotechnology Information (NCBI). Primer design was performed using a commercial program (Prime; GENIUSnet HUSAR; KYE Systems, Heidelberg, Germany) and Primer Premier 5 (PREMIER Biosoft International; Palo Alto, CA). Primers were ordered from a commercial synthesis service (Thermo Electron Corporation, Ulm, Germany). Details are shown in Table 1.

The PCRs were performed in a thermocycler (iCycler; Bio-Rad, Hercules, CA) using a commercial fluorescence detection kit (QuantiTect SYBR Green PCR kit; Qiagen). After an initial heat activation (15 min at 95 °C) 40 cycles of 15 s at 94 °C, 30 s at 59 °C, and 45 s at 72 °C were run with cyclic fluorescence measurements at the end of the annealing phase. The volume of a single reaction added up to 15 μl containing 1 ng template and a final primer concentration of 0.6 μM each. All PCR products were positively verified by automated sequencing. 18S rRNA and β-actin both served as reference genes for semiquantitative analysis.

Data analysis and statistics

Statistical data analysis was based on the threshold cycles (C_T) and performed as described elsewhere [33]. The efficiency (E) of the PCR reaction for each primer pair is determined by dilution series. The efficiency (E) of the PCR reaction for each primer pair is determined by dilution series and the obtained values are as follows: 2.02 for 18S rRNA, 1.99 for β-actin, 1.98 for Egr-1, 1.89 for VIP, and 1.97 for Shh.

Measurements were performed in triplicate and mean values of C_T were used for further analysis. The mean normalized expression (MNE) [34] was used to compare relative expression levels among different groups. A logarithmic transformation of the MNE was used to assure a normal distribution of the residuals for further statistical analysis [34].

18S rRNA and β-actin are both reference genes; Egr-1, VIP, and Shh are target genes. A logarithmic transformation of the MNE was used to assure a normal distribution of the residuals for further statistical analysis.

If multiple comparisons were involved, for example to analyze the influence of different light exposure intervals and to compare different groups with each other, analysis of variance (ANOVA) was carried out (1) to analyze the influence of different light exposure intervals on the mRNA and protein expression of the individual groups (control groups, diffuser treated groups, ND filter treated groups), (2) to analyze the influence of light/dark conditions on Egr-1 protein expression, and (3) to analyze the mRNA expression differences between the groups treated with different ND filters for the same time period. A significant ANOVA was followed by a Tukey-Kramer test for post-hoc analysis.

For the analysis of paired data (diffuser compared to ND filter treated eyes of single animals), paired t-tests were applied and results were Bonferroni corrected for multiple comparisons.

Influence of treatment and light exposure on reference gene expression

The independence of reference gene expression under treatment conditions is essential for semiquantitative analyses. By plotting the C_T values of 18S rRNA and β-actin of all groups and performing an ANOVA, a significant influence of treatment and/or time could be found for both genes. Nineteen individual groups were included in the ANOVA, overall sample C_T mean was 20.13±0.32 for β-actin (lowest C_T mean 19.85, highest C_T mean 20.49) and 9.84±0.57 for 18S rRNA (lowest C_T mean 9.21, highest C_T mean 10.99). Even though the ANOVA for β-actin showed significant expression changes, the variation was rather small. Additionally, similar results were obtained when statistical analyses were done using only the raw data (C_T values) without normalization to a reference gene.

Another measurement for the variability of a PCR is the calculation of the coefficient of variation (CV). The CV is determined by dividing the standard deviation (SD) of each triplicate by the appropriate C_T mean. For 18S rRNA the CV was 1.14%±0.57, whereas for β-actin it was 0.52%±0.29.

Based on these results, β-actin was used as a housekeeping gene for further data analysis.

Correlation of gene expression in control animals

The mRNA expression of Egr-1 (p<0.001), VIP (p<0.001) and Shh (p<0.01) was found to be correlated in both eyes of untreated animals, therefore expression data were pooled and the mean was used for further comparisons.

Results

Changes of mRNA expression in control animals over time

The mRNA expression levels of Egr-1, VIP, and Shh were measured at different points during the 12 h light period, and 30 min after light offset. Significant changes were observed for all genes (Figure 3).

Egr-1 mRNA expression was strikingly upregulated immediately after light onset with a peak after 30 min (20.5 fold increase compared to the initial value at 0 min; Tukey-Kramer test, p<0.001). After 6 h, the mRNA level of Egr-1 had returned to the initial value. Towards the end of the twelve hour light cycle, Egr-1 mRNA level increased again (6 h light compared to 12 h light, Tukey-Kramer test, p<0.05). A further increase of Egr-1 mRNA expression was observed 30 min after light offset (4.2 fold increase compared to the value at the end of the light phase; Tukey-Kramer test, p<0.001). The mRNA levels of Egr-1 in the retina were similarly high 30 min after the dark to light transition and the light to dark transition (Figure 3A).

VIP mRNA expression increased significantly during the second half of the 12 h light period (2.8 fold increase from 6 h light to 12 h light; Tukey-Kramer test, p<0.001). The beginning of the dark period did not significantly influence VIP mRNA expression (Figure 3B).

The expression pattern of Shh mRNA resembled the pattern observed for VIP mRNA. After an initial decrease 2 h after light onset (1.6 fold decrease compared to the initial value at 0 min; Tukey-Kramer test, p<0.01), mRNA expression increased steadily over the next 10 h (2 fold increase in mRNA expression; Tukey-Kramer test, p<0.001). Expression of Shh mRNA, like that of VIP mRNA was not influenced by the short dark period at the end of the day (Figure 3C).

Distribution of Egr-1 labeled cells in the retina of untreated mice

To gain more information on the density of Egr-1 protein was labeled by Egr-1 antibody in retinal cells labeled amacrine cells (AC), bipolar cells (BC) and cells of the ganglion cell layer (GCL) were counted separately in the center and the periphery. The number of Egr-1 expressing neurons was counted in 6 untreated animals one hour after light onset in the morning. As a result, the cell density analysis of all three cell types (AC, BC, and GCL) showed no significant difference between periphery and center (Figure 4).

Changes in Egr-1 protein expression in control animals over time

The diurnal regulation of Egr-1 protein was separately analyzed in AC, BC, and the GCL. The animals either received no light, 1, 4, or 6.5 h of light, or they were kept in darkness for one hour following the regular 12 h light cycle (Figure 5).

The number of Egr-1 immunoreactive cells displayed prominent changes, increasing after light onset in all three cell types (Tukey-Kramer test, AC: p<0.001; Tukey-Kramer test, GCL: p<0.001; Tukey-Kramer test, BC: p<0.05) and subsequently returning to baseline expression levels after 6.5 h. After light offset, a significant increase of Egr-1 expression was found in the bipolar cells (compared to the initial value at 0 min, Tukey-Kramer test, p<0.05). Since Egr-1 was equally expressed in the center and the periphery, cell counts across the retinal section were averaged (cell count/microscope fields).

Influence of different light conditions on Egr-1 protein expression

More refined light conditions were applied to understand whether the regulation of Egr-1 protein is mainly regulated by light/dark transitions or by the time of day. The conditions are listed in Table 2. ANOVA followed by Tukey-Kramer test was applied for each cell type to analyze differences between all groups (1-6).

In the AC (Figure 6A) and the GCL (Figure 6C), it can be seen that Egr-1 protein expression primarily depends on the illumination condition that was experienced just prior to the preparation. The number of Egr-1 immunoreactive amacrine cells was not different among the groups whose final experience was darkness (groups 1-3), whereas Egr-1 protein expression was significantly higher in all groups that were sacrificed after experience of light (Tukey-Kramer test, p<0.001). Within the groups that were recently exposed to light (groups 4-6), the number of Egr-1 immunoreactive amacrine cells was significantly higher in animals that were kept in darkness for a longer period of time when compared to the two other groups (group 5 compared to group 4 and group 5 compared to group 6, Tukey-Kramer test, p<0.01).

In the GCL, Egr-1 protein expression was also significantly different between groups 1-3 and groups 4-6 (Tukey-Kramer test, p < 0.001), but the duration of the dark phase (group 5) did not play a role in the regulation of protein expression.

In the bipolar cells, Egr-1 protein expression was highest in animals that were kept in darkness until the early afternoon (group 5). The expression level of this group differed significantly from that from groups 1-4 (Tukey-Kramer test, p<0.001). The recently experienced light condition seemed to have only a minor influence on protein expression, since the only significant difference occurred between groups 1 and 6 (Tukey-Kramer test, p<0.01).

Effects of bilateral treatment with neutral density (ND) filters on mRNA concentrations

To investigate the effect of changes in retinal illumination on mRNA expression, animals were bilaterally fitted with ND filters of differing light attenuation and with completely black diffusers. All animals were exposed to light for 30 min in the morning (Figure 7).

Based on an average illumination of 120 lux, 0.1 log unit ND filters reduced the light intensity to 96 lux, 0.3 log unit ND filters to 60 lux, and 0.5 log unit ND filters to 38 lux.

Gradual attenuation of retinal image illumination by ND filters had no significant effect on Egr-1 mRNA levels. Interestingly, Egr-1 mRNA expression increased in the animals wearing black diffusers at the beginning of the light phase (5.2 fold increase compared to the value at the end of the dark phase as shown in Figure 3), even though the expression level remained far below the control group that was exposed to light for 30 min (3.9 fold difference, Tukey-Kramer test, p<0.001; Figure 7A).

Neutral density filter wear did not influence VIP mRNA levels (Figure 7B), while the Shh mRNA level also increased significantly in animals wearing black diffusers (1.6 fold increase compared to the group treated with 0.1 log unit ND filters, Tukey-Kramer test, p<0.05 and 1.4 fold increase compared to the group treated with 0.3 log unit ND filters, Tukey-Kramer test, p<0.05; Figure 7C).

Regulation of mRNA levels by retinal image quality

The visual control of eye growth requires good focus and, accordingly, spatial features in the retinal image are analyzed. To gain information about the influence of retinal image quality on the regulation of Egr-1, VIP, and Shh in mice, the expression of their mRNA levels and Egr-1 protein was studied in eyes treated with frosted diffusers and intensity matched ND filters.

Over time, the mRNA expression of Egr-1 varied significantly in the eyes treated with frosted diffusers and ND filters, as was determined by ANOVA. The time course was similar to animals with normal visual experience, with a peak in expression 30 min after the light was turned on (Figure 8A). Even though the Egr-1 mRNA expression seemed to be lower in the eyes treated with diffusers when compared to their intensity matched ND filter treated fellow eyes, the analysis by paired t-tests revealed no significant difference Figure 8A.

The VIP mRNA content was found to be increased in the eyes of animals treated for 120 min, but statistically, the increase was significant only in the ND filter treated eyes (2.1 fold increase compared to both the 30 and 360 min groups, Tukey-Kramer test, p<0.01; Figure 8B). It is nevertheless clear that the ND filter and diffuser treated eyes track together, but that the variance in the ND filter treated groups is lower in most cases. An effect specific to the ND filter, therefore, seems unlikely.

In contrast, retinal Shh mRNA levels were not affected by treatment with diffusers or intensity matched ND filters (Figure 8C).

Regulation of Egr-1 protein expression by image quality

Histological analyses were performed to assess changes of Egr-1 protein expression separately in AC, BC, and GCL after the treatment with diffusers and ND filters. Paired t-tests were applied to analyze changes in Egr-1 protein expression. In AC, Egr-1 expression was similar in eyes treated with diffusers and eyes treated with intensity matched ND filters (Figure 9A). In the BC, Egr-1 expression was decreased in the diffuser treated eyes with blurred vision, compared to the ND filter treated fellow eyes with normal vision (paired t-test, p<0.05; Figure 9B). A similar result was found for the cells in the GCL (paired t-test, p<0.05; Figure 9C).

Egr-1 mRNA expression after treatment with diffusers and ND filters in a visually refined environment

Instead of the normal cage environment that was used for all previous experiments, mice were treated with diffusers on one eye and ND filters on the contralateral eye and were then put in a rotating drum for 30 min. Additionally, animals were exposed to light for 5.5 h before the experiment to assure a low baseline expression of Egr-1 mRNA. The stripe pattern in the drum consisted of 0.1 cycles per deg. Differences in mRNA expression between the diffuser and ND filter treated eyes were evaluated by a paired t-test. Egr-1 mRNA levels were found to be significantly higher in the eyes treated with ND filters compared to the diffuser treated eyes (1.45 fold difference, paired t-test, p<0.05; Figure 10). This suggests that Egr-1 expression might be controlled by the spatial features in the retinal image.

Discussion

Regulation of Egr-1 mRNA and protein expression by light

We have found a very rapid and pronounced induction of Egr-1 mRNA expression both after light to dark and dark to light transitions in the C57BL/6 mouse retina.

Egr-1 mRNA expression has previously been investigated in the retinal degeneration slow (Rds) mutant mouse and the corresponding BALB/c control mouse, where the level of Egr-1 mRNA was upregulated after light offset with a peak 2 h into the dark. Contrary to our present findings, only a minor increase in Egr-1 mRNA was found in these animals after light onset [35]. One possible explanation for the differences observed between those two mouse strains might be the melatonin deficiency of C57BL/6 mice [36]. The lack of melatonin in the C57BL/6 mice might lead to altered gene expression in some retinal cells or the whole retina, even though it was shown that melatonin does not play a crucial role in the generation or maintenance of circadian rhythms [37,38].

The expression patterns of the Egr-1 protein and the mRNA were similar after light onset, but the induction of protein was generally less pronounced and only observed in the bipolar cells after light offset. A transient increase of Egr-1 mRNA was already found in the retina of rats that were exposed to room light from the dark. It seems as though this augmentation is associated with the activation of cholinergic, nicotinic, muscarinic, and NMDA receptors, since it can be suppressed by the corresponding receptor antagonists [39].

The strong influence of light on the regulation of Egr-1 protein is supported by the experiment with different light and dark cycles. These experiments suggest that the recent history of light stimulation is crucial for the Egr-1 expression level, at least in the amacrine cells and the Egr-1 immunoreactive cells in the ganglion cell layer. Such distinct, cell specific changes in expression have been described before for the immediate early gene c-fos gene in the rat retina [40].

The reason for the striking upregulation of Egr-1 protein in the amacrine cells and the bipolar cells after a prolonged darkness period of 17.5 h, followed by one hour of light, is currently not understood but points to an additional circadian influence on Egr-1 protein expression. In a recent study it was proposed that the induction of immediate early genes (like Egr-1) in the rodent retina is controlled by circadian inputs [40]. Circadian regulation might also explain why Egr-1 mRNA expression was increased in animals wearing completely black diffusers. Another explanation for the measured increase in Egr-1 mRNA with black diffusers might be the change in locomotor activity, which is associated with the transfer of the mice from darkness to light, or that very small amounts of light pass through the rim of the black diffusers. Anesthesia and preparation steps cannot account for the increase in Egr-1 mRNA expression since all groups were prepared under the same conditions and the control group without previous light exposure did not show an increased Egr-1 expression. Moreover, a light sensing function of the pineal organ as described in neonatal rats [41] seems unlikely in young mice.

A light stimulated increase in Egr-1 expression by light has already been demonstrated in a variety of tissues in the central nervous system. Egr-1 was found in the suprachiasmatic nuclei (SCN) of rats, hamsters and mice where its expression is induced by photic stimulation during the subjective night [42-45]. The SCN of the hypothalamus is considered as a circadian pacemaker. The cell activity within the SCN can be altered by synchronization of the circadian rhythm to lighting cycles. However, the function of Egr-1 in this tissue is yet unknown. In the visual cortex of rats and mice, Egr-1 expression was found to decrease after several days of dark adaptation. Its expression was rapidly restored to normal levels upon light exposure. Likewise, Egr-1 expression in the visual cortex can be reduced dramatically by blocking afferent visual activity by tetrodotoxin [46].

The regulation of Egr-1 expression seems to be similar in chicks and mice. In chicks, the duration of light exposure influences ZENK (the avian Egr-1 homolog) expression where the number of ZENK-expressing amacrine and bipolar cells varied over the day, increasing after light onset in the morning and declining over the subsequent hour [14].

Regulation of VIP and Shh transcript levels by light

The changes of VIP mRNA expression in the mouse retina are similar to previously found patterns of VIP mRNA in the rat SCN [47], where expression was low during the day and higher towards the end of the light period.

VIP is also expressed in the SCN of mice [48] and an involvement of VIP in circadian rhythmicity has been proposed before. Diurnal regulation of the vasoactive intestinal peptide [49] and its mRNA [50] in the rat SCN have been described earlier.

The expression pattern of Shh mRNA strongly resembled the pattern of VIP mRNA, with an initial decrease followed by a constant increase towards the end of the light period. The regulation of Shh mRNA over time has not yet been described in the mouse eye.

Effects of spatial features in the retinal image on Egr-1 expression

A link between Egr-1 and image focus was first described in the chicken where ZENK (Egr-1) is regulated according to the sign of imposed defocus in a subset of amacrine cells [14].

In mice, a significant decrease (32%) of Egr-1 mRNA in eyes treated with diffusers compared to eyes treated with intensity matched ND filters could be found after a total light exposure of 6 h, whereof the animals were treated for 30 min and spent the treatment period in a rotating drum that was inside covered with a stripe pattern of high contrast. The special experimental procedure was designed to circumvent the heavy increase of Egr-1 mRNA that is usually observed in the morning when the animals are exposed to light for the first time after the night cycle. In the afternoon, the Egr-1 mRNA level is low and it is more likely to detect minor changes that might be due to the altered visual experience. A reason why no effect was detected after 15 or 30 min treatment in the morning might be that the strong light driven increase of Egr-1 mRNA expression obscured the potentially minor influence of the image contrast on the Egr-1 expression.

Additionally, Egr-1 protein expression in mice was downregulated only in a subset of retinal cells (BC and cells of the GCL) of diffuser treated eyes. Such a subtype specific regulation was also found in chicks and macaque retina, and total mRNA extraction might cover most of the cell type specific effects.

Interestingly, Egr-1 mRNA expression was not affected by differences in retinal image illumination when the eyes were bilaterally treated with ND filters, whereas Egr-1 expression decreased when mice were treated with a diffuser on one eye and a ND filter on the other eye. Since Egr-1 mRNA expression is not altered by bilateral ND filter treatment but by unilateral diffuser treatment with similar light attenuation, a contralateral eye effect might be involved.

Altogether, these findings suggest two different visual controls for Egr-1 expression in the mouse retina: (1) light/dark transitions have a major effect on Egr-1 expression, whereas absolute differences in retinal illumination play no or only a minor role and (2) spatial frequency and contrast changes seem to have a minor, but significant, influence on both Egr-1 mRNA and protein expression.

The role of Egr-1 for the visual function of the mouse eye will be studied in more detail in the future. Acting as a transcription factor, Egr-1 can regulate the expression of a number of other genes, including growth factors. A role for NGFI-A (Egr-1) in the regulative mechanisms of retinal plasticity was already shown in rat [51] and primate [52].

Effects of image quality and retinal illumination on VIP and Shh mRNA levels

In the chick, intravitreal injection of porcine VIP led to a decrease of form deprivation myopia development in a dose dependent manner [22]. In the mouse, neither visual form deprivation nor changes in retinal illumination did induce changes in VIP mRNA expression. A reason therefore might be the relatively short treatment period, since chicks were treated for longer time periods (8 days). Changes in retinal illumination did not influence VIP mRNA levels.

Shh mRNA expression was shown to increase in chicks that develop form deprivation myopia [26]. In the mouse, expression of Shh mRNA (like VIP mRNA) was not influenced by deprivation of sharp images, although its regulation was altered over the day. A reason might again be the short treatment period. Changes in retinal illumination induced by ND filters did not influence Shh mRNA level, whereas the treatment with black diffusers led to an increased expression, possibly for the same reasons as described for Egr-1.

Summary

Among the three genes that were previously proposed to be involved in the visual control of eye growth in different animal models, and that were studied here, Egr-1 was most responsive to visual stimulation in the retina of mice. The retinal expression of Egr-1 mRNA and protein was strongly regulated by light, and in particular by dark to light and light to dark transitions. Furthermore, both Egr-1 mRNA and protein levels seem to be regulated by retinal image quality with lower expression levels in diffuser treated compared to the attenuation matched ND filter treated eyes. We therefore hypothesize that Egr-1 in mice might be expressed by a focus sensitive subpopulation of retinal neurons, that could be analog to the glucagonergic amacrine cells in chicks and GAD65-immunoreactive cells in the macaque retina. It nevertheless remains unclear, whether Egr-1 modulates eye growth and refractive development in mice and further studies are necessary to understand the role of Egr-1 in the murine retina.

VIP and Shh mRNA concentrations were also regulated by light and the time of day but none of those two genes was influenced by changes in retinal image quality. It could well be that they are, in fact, not influenced by form deprivation, or that the treatment periods were too short, or that the illumination during the experiments was too low.

Acknowledgements

The authors thank the AG Wissinger for sequence analysis of the PCR products and Dr. Klaus Dietz for the statistical advice. This study was supported by the DFG FE 450/1-1.

References

1. Wallman J. Retinal control of eye growth and refraction. Prog Ret Eye Res 1993; 12:133-53.

2. Schaeffel F, Burkhardt E. Measurement of refractive state and deprivation myopia in a black wildtype mouse. ARVO Annual Meeting; 2002 May 5-10; Fort Lauderdale (FL).

3. Schaeffel F, Burkhardt E, Howland HC, Williams RW. Measurement of refractive state and deprivation myopia in two strains of mice. Optom Vis Sci 2004; 81:99-110.

4. Sukhatme VP, Cao XM, Chang LC, Tsai-Morris CH, Stamenkovich D, Ferreira PC, Cohen DR, Edwards SA, Shows TB, Curran T, Le Beau MM, Adamson ED. A zinc finger-encoding gene coregulated with c-fos during growth and differentiation, and after cellular depolarization. Cell 1988; 53:37-43.

5. Gashler A, Sukhatme VP. Early growth response protein 1 (Egr-1): prototype of a zinc-finger family of transcription factors. Prog Nucleic Acid Res Mol Biol 1995; 50:191-224.

6. Cao XM, Koski RA, Gashler A, McKiernan M, Morris CF, Gaffney R, Hay RV, Sukhatme VP. Identification and characterization of the Egr-1 gene product, a DNA-binding zinc finger protein induced by differentiation and growth signals. Mol Cell Biol 1990; 10:1931-9.

7. Day ML, Fahrner TJ, Aykent S, Milbrandt J. The zinc finger protein NGFI-A exists in both nuclear and cytoplasmic forms in nerve growth factor-stimulated PC12 cells. J Biol Chem 1990; 265:15253-60.

8. Waters CM, Hancock DC, Evan GI. Identification and characterisation of the egr-1 gene product as an inducible, short-lived, nuclear phosphoprotein. Oncogene 1990; 5:669-74.

9. Fu M, Zhu X, Zhang J, Liang J, Lin Y, Zhao L, Ehrengruber MU, Chen YE. Egr-1 target genes in human endothelial cells identified by microarray analysis. Gene 2003; 315:33-41.

10. Wang ZY, Deuel TF. An S1 nuclease-sensitive homopurine/homopyrimidine domain in the PDGF A-chain promoter contains a novel binding site for the growth factor-inducible protein EGR-1. Biochem Biophys Res Commun 1992; 188:433-9.

11. Khachigian LM, Lindner V, Williams AJ, Collins T. Egr-1-induced endothelial gene expression: a common theme in vascular injury. Science 1996; 271:1427-31.

12. Hu RM, Levin ER. Astrocyte growth is regulated by neuropeptides through Tis 8 and basic fibroblast growth factor. J Clin Invest 1994; 93:1820-7.

13. Kim SJ, Glick A, Sporn MB, Roberts AB. Characterization of the promoter region of the human transforming growth factor-beta 1 gene. J Biol Chem 1989; 264:402-8.

14. Fischer AJ, McGuire JJ, Schaeffel F, Stell WK. Light- and focus-dependent expression of the transcription factor ZENK in the chick retina. Nat Neurosci 1999; 2:706-12.

15. Zhong X, Ge J, Smith EL 3rd, Stell WK. Image defocus modulates activity of bipolar and amacrine cells in macaque retina. Invest Ophthalmol Vis Sci 2004; 45:2065-74.

16. Said SI, Mutt V. Isolation from porcine-intestinal wall of a vasoactive octacosapeptide related to secretin and to glucagon. Eur J Biochem 1972; 28:199-204.

17. Longshore MA, Makman MH. Stimulation of retinal adenylate cyclase by vasoactive intestinal peptide (VIP). Eur J Pharmacol 1981; 70:237-40.

18. Schorderet M, Sovilla JY, Magistretti PJ. VIP-and glucagon-induced formation of cyclic AMP in intact retinae in vitro. Eur J Pharmacol 1981; 71:131-3.

19. Tornqvist K, Loren I, Hakanson R, Sundler F. Peptide-containing neurons in the chicken retina. Exp Eye Res 1981; 33:55-64.

20. Tornqvist K, Uddman R, Sundler F, Ehinger B. Somatostatin and VIP neurons in the retina of different species. Histochemistry 1982; 76:137-52.

21. Stone RA, Laties AM, Raviola E, Wiesel TN. Increase in retinal vasoactive intestinal polypeptide after eyelid fusion in primates. Proc Natl Acad Sci U S A 1988; 85:257-60.

22. Seltner RL, Stell WK. The effect of vasoactive intestinal peptide on development of form deprivation myopia in the chick: a pharmacological and immunocytochemical study. Vision Res 1995; 35:1265-70.

23. Echelard Y, Epstein DJ, St-Jacques B, Shen L, Mohler J, McMahon JA, McMahon AP. Sonic hedgehog, a member of a family of putative signaling molecules, is implicated in the regulation of CNS polarity. Cell 1993; 75:1417-30.

24. Jensen AM, Wallace VA. Expression of Sonic hedgehog and its putative role as a precursor cell mitogen in the developing mouse retina. Development 1997; 124:363-71.

25. Moshiri A, Reh TA. Persistent progenitors at the retinal margin of ptc+/- mice. J Neurosci 2004; 24:229-37.

26. Escano MF, Fujii S, Sekiya Y, Yamamoto M, Negi A. Expression of Sonic hedgehog and retinal opsin genes in experimentally-induced myopic chick eyes. Exp Eye Res 2000; 71:459-67.

27. Akamatsu S, Fujii S, Escano MF, Ishibashi K, Sekiya Y, Yamamoto M. Altered expression of genes in experimentally induced myopic chick eyes. Jpn J Ophthalmol 2001; 45:137-43.

28. Grimm C, Wenzel A, Hafezi F, Reme CE. Gene expression in the mouse retina: the effect of damaging light. Mol Vis 2000; 6:252-60 <http://www.molvis.org/molvis/v6/a34/>.

29. Bartmann M, Schaeffel F. A simple mechanism for emmetropization without cues from accommodation or colour. Vision Res 1994; 34:873-6.

30. Schmucker C, Seeliger M, Humphries P, Biel M, Schaeffel F. Grating acuity at different luminances in wild-type mice and in mice lacking rod or cone function. Invest Ophthalmol Vis Sci 2005; 46:398-407.

31. Bitzer M, Schaeffel F. Defocus-induced changes in ZENK expression in the chicken retina. Invest Ophthalmol Vis Sci 2002; 43:246-52.

32. Haverkamp S, Wassle H. Immunocytochemical analysis of the mouse retina. J Comp Neurol 2000; 424:1-23.

33. Buck C, Schaeffel F, Simon P, Feldkaemper M. Effects of positive and negative lens treatment on retinal and choroidal glucagon and glucagon receptor mRNA levels in the chicken. Invest Ophthalmol Vis Sci 2004; 45:402-9.

34. Simon P. Q-Gene: processing quantitative real-time RT-PCR data. Bioinformatics 2003; 19:1439-40.

35. Agarwal N. Diurnal expression of NGF1-A mRNA in retinal degeneration slow (rds) mutant mouse retina. FEBS Lett 1994; 339:253-7.

36. Ebihara S, Marks T, Hudson DJ, Menaker M. Genetic control of melatonin synthesis in the pineal gland of the mouse. Science 1986; 231:491-3.

37. Stehle JH, von Gall C, Korf HW. Organisation of the circadian system in melatonin-proficient C3H and melatonin-deficient C57BL mice: a comparative investigation. Cell Tissue Res 2002; 309:173-82.

38. Stehle JH, von Gall C, Korf HW. Melatonin: a clock-output, a clock-input. J Neuroendocrinol 2003; 15:383-9.

39. Gudehithlu KP, Neff NH, Hadjiconstantinou M. c-fos and NGFI-A mRNA of rat retina: evidence for light-induced augmentation and a role for cholinergic and glutamate receptors. Brain Res 1993; 631:77-82.

40. Humphries A, Carter DA. Circadian dependency of nocturnal immediate-early protein induction in rat retina. Biochem Biophys Res Commun 2004; 320:551-6.

41. Blackshaw S, Snyder SH. Developmental expression pattern of phototransduction components in mammalian pineal implies a light-sensing function. J Neurosci 1997; 17:8074-82.

42. Kilduff TS, Vugrinic C, Lee SL, Milbrandt JD, Mikkelsen JD, O'Hara BF, Heller HC. Characterization of the circadian system of NGFI-A and NGFI-A/NGFI-B deficient mice. J Biol Rhythms 1998; 13:347-57.

43. Rusak B, McNaughton L, Robertson HA, Hunt SP. Circadian variation in photic regulation of immediate-early gene mRNAs in rat suprachiasmatic nucleus cells. Brain Res Mol Brain Res 1992; 14:124-30.

44. Rusak B, Robertson HA, Wisden W, Hunt SP. Light pulses that shift rhythms induce gene expression in the suprachiasmatic nucleus. Science 1990; 248:1237-40.

45. Sutin EL, Kilduff TS. Circadian and light-induced expression of immediate early gene mRNAs in the rat suprachiasmatic nucleus. Brain Res Mol Brain Res 1992; 15:281-90.

46. Worley PF, Christy BA, Nakabeppu Y, Bhat RV, Cole AJ, Baraban JM. Constitutive expression of zif268 in neocortex is regulated by synaptic activity. Proc Natl Acad Sci U S A 1991; 88:5106-10.

47. Ban Y, Shigeyoshi Y, Okamura H. Development of vasoactive intestinal peptide mRNA rhythm in the rat suprachiasmatic nucleus. J Neurosci 1997; 17:3920-31.

48. Cassone VM, Speh JC, Card JP, Moore RY. Comparative anatomy of the mammalian hypothalamic suprachiasmatic nucleus. J Biol Rhythms 1988; 3:71-91.

49. Takahashi Y, Okamura H, Yanaihara N, Hamada S, Fujita S, Ibata Y. Vasoactive intestinal peptide immunoreactive neurons in the rat suprachiasmatic nucleus demonstrate diurnal variation. Brain Res 1989; 497:374-7.

50. Okamoto S, Okamura H, Miyake M, Takahashi Y, Takagi S, Akagi Y, Fukui K, Okamoto H, Ibata Y. A diurnal variation of vasoactive intestinal peptide (VIP) mRNA under a daily light-dark cycle in the rat suprachiasmatic nucleus. Histochemistry 1991; 95:525-8.

51. Pinaud R, Tremere LA, Penner MR, Hess FF, Barnes S, Robertson HA, Currie RW. Plasticity-driven gene expression in the rat retina. Brain Res Mol Brain Res 2002; 98:93-101.

52. Pinaud R, De Weerd P, Currie RW, Fiorani Jr M, Hess FF, Tremere LA. Ngfi-a immunoreactivity in the primate retina: implications for genetic regulation of plasticity. Int J Neurosci 2003; 113:1275-85.