Relative Axial Myopia in Egr-1 (ZENK) Knockout Mice

Abstract

purpose. Experiments in chickens have implicated the transcription factor ZENK (also known as Egr-1, NGFI-A, zif268, tis8, cef5, and Krox24) in the feedback mechanisms for visual control of axial eye growth and myopia development. ZENK is upregulated in retinal glucagon amacrine cells when axial eye growth is inhibited by positive spectacle lens wear and is downregulated when it is enhanced by negative spectacle lens wear, suggesting that ZENK may be linked to an inhibitory signal for axial eye growth. This study was undertaken to determine whether a Egr-1 ^−/− knockout mouse mutant, lacking ZENK completely, has longer eyes and more myopic refraction, than do Egr-1 ^{+/ −} heterozygous and Egr-1 ^+/+ wild-type mice with near-identical genetic backgrounds.

methods. Eye growth and refractive development were tracked from day P28 to P98. Corneal radius of curvature was measured with infrared photokeratometry, refractive state with infrared photoretinoscopy, and ocular dimensions with low-coherence interferometry. As a functional vision test, grating acuity was determined in an automated optomotor task. The abundance of ZENK protein in the retina was quantified by immunohistochemistry.

results. Egr-1 knockout mice had longer eyes and a relative myopic shift in refraction, with additional minor effects on anterior chamber depth and corneal radius of curvature. Paraxial schematic eye modeling suggested changes in the optics of the crystalline lens as well. With increasing age, the differences between mutant and wild-type mice declined, although the differences in refraction persisted over the observation period. Grating acuity was not affected by the lack of the Egr-1 protein during development.

conclusions. Although it has been shown that different mouse strains may have differently large eyes, the present study shows that a specific gene knockout can produce relative myopia, compared with the wild-type with near-identical genetic background. Further experiments are needed to determine whether the observed effects of Egr-1 deletion are due to changes in function within the retina or other ocular tissues or to changes of function in other systems that may affect ocular growth from outside the eye.

In vertebrate animal models, it has been shown that refractive development and the axial growth of the eye are under visual control. ¹ Messengers released from the retina affect the expansion of the different fundal layers of the eye in a highly specific way. Development of myopia, induced by negative-power lenses or frosted occluders, is accompanied by choroidal thinning and an increase in scleral proteoglycan synthesis in chicks and tree shrews ^{2 3} and increased proliferation of chondrocytes in chicks, ⁴ whereas the opposite is true of hyperopia induced by positive-power lenses. The identification of regulatory molecular targets for emmetropization in the fundal layers of the eye has been a research topic for several years. Still, we are far from understanding the complete network of biochemical interactions in the signaling cascade. 

The most intensively studied retinal messengers in the chicken model of myopia are dopamine, ^{5 6} cholinergic antagonists, ⁷ retinoic acid, ⁸ and glucagon. ^{9 10} Intravitreal application of glucagon agonists inhibits the development of myopia, whereas injection of a glucagon antagonist inhibits the development of hyperopia. ^{10 11 12} Different from the chicken, glucagon-containing amacrine cells are not found in the retina in mice, ¹³ although the glucagon receptor mRNA has been detected. ¹⁴ In rat retina, binding assays indicate the presence of receptor-like, high-affinity glucagon-binding sites. ¹⁵ Some other members of the glucagon superfamily were found, although these proteins do not seem to be visually regulated. ¹³ Therefore, the role of glucagon in the visual control of eye growth in mammals is uncertain. Other peptides, perhaps also from the glucagon superfamily, may have taken over its role. 

Recently, microarray studies of retinal transcripts were initiated to identify new genes associated with eye-growth regulation in mice ¹⁶ (Brand C et al. IOVS 2006:ARVO E-Abstract 3327). In addition, microarray analyses of gene expression levels in Egr-1 knockout mice are under way. Comparison between the results of both studies will lead to a better understanding of the regulatory pathways and signaling cascades, particularly regarding Egr-1 and its target genes. 

However, the glucagon amacrine cell in the chick is interesting for another reason: It expresses the transcription factor ZENK in tight correlation with the sign of defocus of the projected image on the retina. ^{17 18} ZENK is upregulated when axial eye growth is inhibited and is downregulated when it is enhanced. Regulation of retinal ZENK by the spatial information in the retinal image was confirmed in mice, at both the mRNA and protein levels (herein called Egr-1, early growth response protein-1, the mammalian orthologue to the avian ZENK). Egr-1 mRNA and Egr-1 protein were found to be reduced in retinas of mouse eyes that had been covered with frosted occluders for only 1 hour, compared with the fellow eyes that were treated with attenuation-matched neutral-density filters. ¹⁹ Furthermore, Egr-1 expression was lower in GAD65 expressing amacrine cells of the rhesus monkey under conditions that induce myopia and higher during conditions that induce recovery from myopia. ²⁰  

Egr-1 is a protein with three zinc-finger domains for sequence specific DNA-binding that belongs to the immediate early gene family. It induces the expression of different growth factors, as for instance platelet-derived growth factor (PDGF)-A chain, ²¹ PDGF-B chain, ²² basic fibroblast growth factor (bFGF), ²³ and transforming growth factor (TGF)-β. ²⁴ Moreover, among others, adhesion molecules (e.g., collagen1A2), ²⁵ cytokines (tumor necrosis factor), ²⁶ and hormones are targets of the transcription factor Egr-1. Egr-1 is also involved in cell proliferation, ^{27 28} macrophage differentiation, ²⁹ synaptic activation and long-term potentiation, ³⁰ has a role as key mediator of inflammation and apoptosis in vascular cells ³¹ and activates the transcription of target genes that provide the products necessary for mitogenesis and differentiation. ^{32 33}  

In summary, ZENK/Egr-1 is an interesting factor in the signaling cascade for the visual control of eye growth. If the simple correlation between the sign of imposed defocus and ZENK expression were more general, it could be expected that mice lacking Egr-1 would develop longer eyes because they lack an important element in the inhibitory circuitry for axial eye growth. We tested this hypothesis in a knockout mouse model, with complete absence of Egr-1. 

Methods

Animals

Experiments were conducted in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and were approved by the University Commission for Animal Welfare (reference AK 3/03). Egr-1 knockout mice, generated on C57/BL6 background, were obtained from Taconic Farms (Germantown, NY) and bred in the university animal facilities after the breeding permit was paid for. Genotyping was performed by PCR with primer sequences supplied by Taconic. In addition, immunohistochemistry for the expression of the ZENK protein in the retina was used at the end of the experiments to confirm correct genotyping. Animals were housed in standard mouse cages with their littermates in a 12-hour light–dark cycle, with free access to water and food pellets. Illuminance in the cages was approximately 200 lux (provided by neon lamps), which is approximately 1000 lux below the illumination levels that induce retinal degeneration in mice. ³⁴ A total of 64 mice were used in the study, 20 Egr-1 ^−/− knockout mice, 27 Egr-1 ^+/− heterozygous mice, and 17 Egr-1 ^+/+ wild-type mice. Measurement sessions, repeated every other week during the experimental period from day P28 to P98, were performed on 1 day in the following order: First, visual acuity was tested at the four different spatial frequencies (0.05, 0.1, 0.2 and 0.4 cyc/deg; a more detailed description is provided later in the article). Subsequently, refractive state was measured. Finally, mice were anesthetized by intraperitoneal injection of a 0.1- to 0.2-mL 1.2% ketamine/1.6% xylazine mixture and body weight and axial length of the eyes were determined. All mice recovered from anesthesia without complication. For organizational reasons, mice could not be always be measured at exactly the same ages. Also, the number of mice varied in the different groups depending on availability (see Table 1 ). Therefore, before analyzing the data sets, we verified that there were no significant differences in the ages of the mice that were included in a data point for an age group (all ANOVAs: P > 0.05). 

Analysis of all animals showed that homozygous mice weighed, in total, approximately 2.2 g less than did the heterozygous and wild-type mice (Tukey-Kramer honest significant difference [HSD] P < 0.01). These differences were most prominent at day 56 (Egr-1 ^{− / −} mice 2.4 g lighter than Egr-1 ^{+/ −} mice; P < 0.05) and day 70 (Egr-1 ^{− / −} mice 2.3 g lighter than Egr-1 ^+/+ mice; P < 0.05 and Egr-1 ^{− / −} mice 4.0 g lighter than Egr-1 ^{+/ −} mice; P < 0.001). Gender had no influence on any of the tested parameters. 

The corneal radius of curvature was measured on another day, because it was not possible to perform all measurements on 1 day. Corneal radius of curvature was measured at the ages of 25, 40, 50, 65, 80, and 95 days. 

Immunohistochemistry

Immunohistochemical labeling of the Egr-1 protein was performed as described by Brand et al., ¹⁹ with a fixation method described by Bitzer and Schaeffel. ¹⁸ Cryostat sections, 12 μm thick, were incubated with a primary antibody for Egr-1 protein (rabbit polyclonal antibody at 1:1250, Egr-1 (588) sc-110, lot number L239; Santa Cruz Biotechnology Inc., Santa Cruz, CA), followed by a secondary antibody, indocarbocyanine (Cy3)-linked rabbit anti-mouse IgG (GE Healthcare, Freiburg, Germany). 

Measurement of the Corneal Radius of Curvature

Infrared photokeratometry was used without cycloplegia or anesthesia, as described by Schmucker and Schaeffel. ³⁵ In brief, we placed the mice on a wooden platform and only slightly restrained them by holding their tails. Eight light-emitting diodes arranged in a circle created eight Purkinje images on the corneal surface. An infrared light–sensitive video camera provided a highly magnified image of the corneal light reflexes, and an image-processing program written by one of us (FS) detected their positions and fitted a circle through their centers at a 25-Hz video frame rate. The radius of the fitted circle was directly proportional to the radius of the circular LED arrangement, the camera distance, and corneal radius of curvature. After calibration of the setup by measurements in five ball bearings with known radii (1.5, 2, 2.5, 3, and 4 mm), the radius of curvature of the cornea could be easily determined by linear extrapolation. Each eye was measured five times, and the averages from both eyes were taken for each animal. The SD of repeated measurements was approximately ±0.02 mm. 

Measurement of Grating Acuity

Grating acuity was determined in an automated optomotor task, as described by Schmucker et al. ³⁶ In brief, mice were individually placed in a cylindrical Perspex container (diameter, 15 cm) that rested on a stationary platform in the center of the optomotor drum (diameter, 63 cm). The drum was rotated at an angular speed of 50 to 60 deg/sec by a DC electromotor. This angular speed has been shown to elicit the best optomotor responses. ³⁶ The optomotor drum was covered inside with vertical square-wave gratings of spatial frequencies of 0.05, 0.1, 0.2, and 0.4 cyc/deg, all presented at close to 100% Michelson contrast. The luminance of approximately 30 cd/m² was generated by a light bulb (60 W). The movements of the mice in the container were automatically recorded from above by a video system. Two variables were analyzed: the average angular speed of the center of mass of the mouse (angular running speed) and angular rotation of the snout–tail body axis (angular orientation speed). Responses of the mouse were defined as the difference between the angular movements measured when the drum was rotating clockwise and those obtained during counterclockwise rotation. To minimize habituation, ³⁷ we reversed the direction of rotation of the optomotor drum every 20 seconds. Reversal was repeated four times for each spatial frequency. Data are provided in degrees per frame, together with their standard deviations, because the movements of the mice were measured from one video frame to the next. Frame rate was 25 Hz. The more the responses differed from 0, the more was the mouse influenced by visual cues. 

Measurement of Refractive State

Refractive state was determined by eccentric infrared photorefraction as described by Schaeffel et al. ³⁸ We placed the mice individually on a platform and slightly restrained them by holding their tails. After aligning their eyes with the camera axis, a video program detected the pupil and determined the refractive state from the slopes of the brightness distribution in the pupil. The conversion from brightness slope into refraction was determined from a prior calibration with trial lenses. ³⁸ As measurements were taken in darkness, and with infrared light, the pupil sizes of the mice were large enough (approximately 2 mm), and no cycloplegia was necessary. Averages of three measurements of refractive state and pupil size were used. 

Measurement of Ocular Dimensions

Axial length and anterior chamber depth were measured (ACMaster; Carl Zeiss Meditec, Jena, Germany). This optical low-coherence interferometer was initially developed for measurements of the anterior chamber in humans but turned out to be very useful also for measurements in mouse eyes. ³⁹ For these measurements, mice had to be slightly anesthetized because the position of their eyes was too variable when they were alert. Because some mice (∼25%) developed irregularities of the corneal surface due to the anesthesia, not all animals could be measured on all occasions. If this was the case, the data of the respective eyes were excluded. Therefore, the number of mice that were measured varied in the different age groups, as shown in more detail in Table 1 . The interferograms of the system (ACMaster; Carl Zeiss Meditec) were analyzed off-line, and only those scans with clear peaks at the anterior corneal surface, the anterior lens surface, and the retinovitreous interface were considered. Means of measurements from both eyes were taken. Axial length was defined as the distance from the anterior corneal surface to the retinovitreous interface, and anterior chamber depth was defined as the distance from the anterior corneal surface to the anterior lens surface. 

Statistics

Because the mice could not be measured in exact 2-week intervals (Table 1) , the data from animals of similar ages were pooled (age bins around 28, 42, 56, 70, 84, and 98 days). The parameter age was then treated as nominal data, like the parameter genotype. Statistical analyses were performed (JMP; SAS Institute Inc., Cary, NC). Data analysis was extended into multiple comparisons only if the preceding multicomparison ANOVA revealed significance of the analyzed parameter. For post hoc tests, the Tukey-Kramer HSD was chosen. 

Significance levels are denoted in the figures by asterisks (*P < 0.05, **P < 0.01, ***P < 0.001; NS, not significant). Asterisks above the bars refer to differences between the homozygous and the wild-type mice, and those below the bars denote differences between the homozygous and the heterozygous mice. 

Results

Immunohistochemical Labeling of Egr-1

Nonspecific labeling was visible in the outer and inner plexiform layers of all mouse retinas. In contrast, specific labeling for Egr-1 was confined to the cell nuclei. As expected, Egr-1 antibodies labeled cell nuclei in the retinas from wild-type (^+/+) and heterozygous (^+/−) mice, but not of homozygous (^−/−) mice (Fig. 1) . The average number of labeled cells, counted over two microscope fields, was 26.7 ± 2.8 and 5.9 ± 1.1 in wild-type and heterozygous mice, respectively (means ± SEM). As previously described by Brand et al., ¹⁹ Egr-1 protein expression was observed in amacrine cells, bipolar cells, and cells in the ganglion cell layer. 

Development of Corneal Radius of Curvature

Multicomparison ANOVA with the parameters genotype and age revealed a highly significant change in corneal radius of curvature with age (P < 0.0001). In the wild-type animals, corneal radius of curvature increased from 1.35 mm at the age of 28 days to 1.54 mm at the age of 98 days (Fig. 2) . After splitting the data into the six age-groups, one-way ANOVAs were performed and revealed a significant effect of genotype on radius of curvature, but only at day 40 (P = 0.0206, Tukey-Kramer post hoc test). At this age, corneal radius of curvature was 0.029 mm smaller in the homozygous mice, compared with wild-type mice (P < 0.05). 

Grating Acuity

Multicomparison ANOVA with the parameters genotype, age, and spatial frequency of the stripes revealed a highly significant influence (P < 0.0001) only of spatial frequency on grating acuity (angular orientation speed – the angular rotation of the snout–tale axis, Fig. 3 ). There was no significant effect of the visual stimulation on the motion of the center of mass of the mice, relative to the drum center (angular running speed, data not shown). No effect of genotype or age on angular orientation speed was found (all four multicomparison ANOVAs, P > 0.05). Apparently, development of normal optomotor grating acuity does not require expression of the Egr-1 protein. The visually elicited responses of the mice in the present study were similar to those in previous studies of another C57J-BL6 wild-type mouse strain. ³⁶  

Refractive State and Pupil Size

Both age and genotype had a significant effect on refractive development (multicomparison ANOVA, P < 0.001). Homozygous Egr-1 knockout mice were less hyperopic than were wild-type mice at all tested ages (Fig. 4 , Tukey-Kramer HSD: days 28, 42, 56, and 98: P < 0.001 each; day 70: P < 0.01; day 84: P < 0.05). A comparison between homozygous and heterozygous mice revealed a significant difference at the ages of 28, 42, 56, and 98 days (Fig. 4 , Tukey-Kramer HSD: P < 0.001 each) and at day 70 (Tukey-Kramer HSD: P < 0.05), but no difference at day 84. 

Pupil size increased with age (multicomparison ANOVA, P < 0.0001) but genotype had little effect (multicomparison ANOVA, P = 0.0882). A minor difference was found only at the age of 28 days, where homozygous mice had slightly smaller pupils than did the wild-type (−0.23 mm; P < 0.05, Tukey-Kramer HSD). 

Ocular Dimensions

As expected, axial eye length increased with age in all groups (Fig. 5A , P < 0.0001). However, also genotype had a significant effect (P < 0.0001). After the data were split into the age-grouped data sets, significant differences between homozygous, heterozygous, and wild-type Egr-1 knockout mice were found at the ages of 42 (P = 0.0003) and 56 (P = 0.0099) days. At 42 days, homozygous knockout mice had longer eyes than did the wild-type mice (+0.059 mm; P < 0.01, Tukey Kramer HSD) and heterozygous mice (+0.051 mm; P < 0.001). At the age of 56 days, this difference remained significant only between the homozygous and the heterozygous knockout mice (+0.040 mm; P < 0.01). The differences in axial eye length between the three genotypes became smaller with age and disappeared at the end of the measurement period. 

Also anterior chamber depth increased continuously with age in all three genotypes (Fig. 5B) . Multicomparison ANOVA on anterior chamber depth with the parameters age and genotype revealed significant influence of age (P < 0.0001) but no influence of genotype (P = 0.1539). After the data were split into age groups, one-way ANOVA showed that the eyes of homozygous mice had deeper anterior chambers than those of heterozygous mice (+14 μm difference) only at the age of 56 days (P < 0.05, Tukey-Kramer HSD). 

Discussion

We describe a mouse mutant in which relative axial myopia develops due to genetic factors. Although genetically determined differences in eye size and eye weight have been described for DBA/2J and C57BL/6J mice, ^{40 41} the present study is the first in which such differences can be attributed to the absence of a single gene product, the Egr-1 protein. This transcription factor is interesting because it is already known to be induced by conditions that affect eye growth in chickens ⁹ and monkeys. ²⁰ Of course, it remains unclear how Egr-1 produces the changes in axial length and refractive error, but at least the genetic background was largely identical with that of the wild-type mice used for comparison. In contrast to previous studies, we measured not only eye weight but also the optical variables (refraction, corneal radius of curvature, axial length, and anterior chamber depth) that were necessary to demonstrate that the eyes had axial myopia. In other studies of mouse mutants (e.g., the nob mouse, showing normal refractive development but increased susceptibility to form-deprivation myopia), only refractive state was analyzed (Fernandes A et al. IOVS 2005:ARVO E-Abstract 2281). 

Effects of the lack of the Egr-1 protein on eye growth were small and were detectable only between the ages of 42 and 56 days. A possible explanation for the temporary effects could be that the gain of the emmetropization feedback loop, partially controlled by Egr-1, declines with age. General eye growth in all directions could have hidden the small effects later in development. 

In the knockout mouse, Egr-1 is absent from every cell in the body, and its absence can be assumed to affect many functions, inside as well as outside the retina. In the sclera, Egr-1 may influence eye growth through affecting extensibility. This possibility will be discussed in more detail later. It could also be that a change of function in the brain, autonomic nervous system, or endocrine system affects ocular growth from outside the eye. For example, Egr-1 expression might normally suppress activity or hormone synthesis and release in the thyroid gland. Then, the absence of Egr-1 could promote ocular elongation through increased availability of thyroid hormone to growing eye tissues. Our observation that homozygous Egr-1 ^−/− mice tend to be of lower weight underlies the assumption that the immediate early gene has several functions throughout the body. In addition, it is well known that other proteins may take over the role of the deleted one in knockout models, ⁴² and this could have happened in our study as well. Therefore, further experiments (for example tissue specific overexpression or inhibition of Egr-1expression in the retina) are necessary to show whether the effects of Egr-1 depletion on eye elongation and refraction are only due to changes in function within the retina or other ocular tissues. 

Dose Effects of Egr-1

The number of Egr-1 immunoreactive amacrine cells was much lower in the heterozygous Egr-1 knockout mice, down to 25%, compared with the wild-type. Despite the lower availability of Egr-1, no differences in axial eye length or refractive state were detected between heterozygous Egr-1 knockout mice and wild-type. Apparently, complete absence of Egr-1 is necessary for temporary axial eye elongation and relative myopia. 

Paraxial Schematic Eye Modeling

The schematic eye for the growing C57BL/6J mouse described by Schmucker and Schaeffel ³⁹ was used to evaluate the optic effects of changes in ocular components. At 42 days of age, an axial elongation of 5.7 μm produces a shift of refraction in the myopic direction by 1 D. Because the eyes of the homozygous knockout mice were 59 μm longer at the age of 42 days (Fig. 5A) , it would be expected that they are approximately 10 D more myopic. In addition, corneal radius declined from approximately 1.45 mm (Fig. 2)to approximately 1.42 mm (difference, 29 μm). With an effective refractive index of 1.332 for modeling the effects of corneal power changes on refractive state, ⁴³ this change should produce another increase of myopia by approximately 5 D. However, the measured difference in refraction was only approximately 5 D in total, which could indicate that the refraction procedure was inaccurate. This is unlikely, however, since the technique was calibrated in mouse eyes with trial lenses. Another explanation could be that there were changes in lens thickness, curvature and power. This assumption is in line with the observation that both axial length differences and corneal curvature difference disappeared with increasing age, while the refraction differences persisted. 

Visual Function without Egr-1

Complete absence of Egr-1 during development did not affect grating acuity in an optomotor task. Given that Egr-1 has an important role in learning and development, ^{30 44} this finding was surprising. The optomotor task requires functional sensory–motor interaction and not only tests retinal function, but is a predominantly reflexive movement that involves no learning. In addition, the depth of focus is approximately ±10 D in the mouse eye. ⁴⁵ The Egr-1 knockout mice are therefore probably able to offset the measured relative myopic shift in the optomotor task. Other behavioral tests may be necessary to detect potential functional deficits in the visual system of the mouse, as the lack of Egr-1 can influence other pathways than the accessory optical subsystem like the retinotectal or the retinogeniculocortical subsystems. 

What Is the Function of Egr-1 in the Visual Control of Eye Growth?

Because Egr-1 knockout mice had a myopic shift in refraction compared with wild-type and heterozygous mice and because the axial length tended to be longer during the first weeks after eye opening, an involvement of Egr-1 in eye growth regulation appears likely. 

One may speculate that a deficiency in Egr-1 leads to a decreased expression of growth factors such as TGF-β and bFGF. The role of these growth factors in ocular growth control has already been suggested, at least in chicks. ^{46 47} Moreover, a possible function of Egr-1 in the control of eye growth could be its regulatory effect on collagen expression in the sclera. The development of high myopia is associated with reduced scleral collagen accumulation, scleral thinning, and loss of scleral tissue, in both humans and animal models. ^{48 49 50} Very recently, Egr-1 was identified as a novel intracellular TGF-β target that is necessary for maximum stimulation of the expression of the type I collagen gene in fibroblasts. ²⁵ In addition, it has been shown that type 1 collagen mRNA is expressed in the sclera of mice. ⁵¹ These two observations make it likely that lack of Egr-1 may also reduce collagen expression in scleral fibroblasts, leading to increased distensibility of the sclera and thereby to more than normal ocular elongation, although the effects on eye growth and on connective tissue in the body do not seem to be large. In wild-type mice, collagen mRNA levels decline with age by a factor of 4 to 5, ¹⁶ which could be one of the reasons that the effect of Egr-1 deficiency on eye growth was only temporary. 

It is not yet known which genes are controlled by Egr-1 at different developmental stages and how these genes regulate eye growth, and therefore microarray analyses on the Egr-1 knockout mouse strain appear promising. The extensive back-crossing of the mice during the experimental period may mean that the genetic background is largely isogenic. Because an Egr-1 ^+/+ wild-type mouse serves as a control, it should be possible to identify novel candidate genes that have inhibitory function on axial elongation of the eye. 

 Supported by the German Research Council Grants DFG FE 415-1/1 and DFG Scha 518/13-1.

 Submitted for publication July 24, 2006; revised August 29, 2006; accepted November 10, 2006.

 Disclosure: R. Schippert, None; E. Burkhard t, None; M. Feldkaempe r, None; F. Schaeffe l, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Table 1.

 

Mean Ages of Mice Measured at Each Time Point

	Genotype	Time Point (d)
				28	42	56	70	84	98
Mice (n)	Homozygous	10	12	12	12	10	11
	Heterozygous	15	26	23	14	18	13
	Wild-type	10	10	10	11	10	10
Mean age (d)	Homozygous	29.0	42.6	55.4	70.4	84.8	99.1
	Heterozygous	28.1	42.1	55.3	70.6	83.7	98.9
	Wild-type	28.4	41.1	55.9	70.8	82.9	100.1

Figure 1.

 

Immunohistochemical labeling for Egr-1 protein in wild-type (^+/+), heterozygous (^+/−), and homozygous (^−/−) retinas. Nonspecific labeling was present in the outer and inner plexiform layers of all mouse retinas. In contrast, specific labeling for Egr-1 was confined to the cell nuclei. Arrows: labeled amacrine cells. The wild-type showed the most abundant Egr-1 labeling but some labeling was also observed in heterozygous mice.

Figure 2.

 

Development of corneal radius of curvature in homozygous, heterozygous, and wild-type Egr-1 knockout mice. Asterisks above the bars refer to differences between the homozygous and the wild-type mice, and those below the bars to differences between the homozygous and the heterozygous mice. The asterisk therefore denotes a significant difference between homozygous and wild-type mice at the age of 40 days (Tukey-Kramer HSD: P < 0.05). There was no significant difference between homozygous and heterozygous mice. Error bars, SEM.

Figure 3.

 

Grating acuity of Egr-1 knockout and wild-type mice, tested in the optomotor drum. The angular orientation speed was tested at the different spatial frequencies (0.05, 0.1, 0.2, and 0.4 cyc/deg). There were no differences in grating acuity between mutant and wild-type mice. Error bars, SEM.

Figure 4.

 

Development of refractive state in homozygous, heterozygous and wild-type mice. *P < 0.05, **P < 0.01, ***P < 0.001; NS, not significant. Asterisks above the horizontal barsindicate significant differences between the homozygous and the wild-type mice, and those below denote significant differences between homozygous and heterozygous mice. Error bars, SEM.

Figure 5.

 

(A) Axial eye growth in homozygous and heterozygous knockout mice, and wild-type. *P < 0.05, **P < 0.01, ***P < 0.001; NS denotes not significant). (B) Growth of anterior chamber depth. Error bars, SEM. Asterisks above the horizontal bar indicate significant differences between homozygous and wild-type mice and those below the bar between homozygous and the heterozygous mice.