Retinal Morphology and ERG Response in the DBA/2NNia Mouse Model of Angle-Closure Glaucoma

Abstract

purpose. To document the time course of retinal dysfunction by scotopic electroretinography (ERG) and by quantitative morphology in eyes of the DBA/2NNia substrain of mouse (DBA) with inherited angle-closure glaucoma.

methods. DBA and control C57BL/6J (C57) mice were studied by ERG recordings from 5 to 15 months of age, and by morphology from 1 to 14 months of age. Scotopic ERGs were simultaneously recorded from both eyes of dark-adapted anesthetized mice. Changes in the central neuronal retina were evaluated by quantitative morphometry performed on serial semithin sections of Epon-embedded eyes.

results. When compared with normal C57 mice, DBA mice showed significant reductions of the a-wave and b-wave amplitudes by 7 to 8 months, and the decline continued as the animals aged. The b-wave implicit time in DBA mice showed a gradual prolongation beginning at 8 months of age, when compared with C57 mice. Logistic regression analyses revealed significant correlations in a- and b-wave amplitude reductions between ipsilateral and contralateral eyes of DBA mice at ages when ERG parameters were greatly altered. Morphologically, thinning of the whole retina was already evident in DBA mice at 4 months of age, but loss of ganglion cells and thinning of the outer plexiform layer were first seen in 7- to 8-month-old animals. These changes progressed to the end of the 13-month period studied.

conclusions. Progressive thinning of the outer retinal layers in DBA mice was found to correlate with decreases in ERG a- and b-wave amplitudes, both occurring from the age of 7 to 8 months onward. Similarities with the findings in human late-stage glaucomatous retinopathy indicate the relevance of this animal model in further glaucoma research.

Recently, there has been increased interest in DBA/2 mice as potential models for secondary angle-closure glaucoma. Abnormalities in the anterior segment of the eye in this strain were first reported in 1986 ¹ and were later followed up by a general pathologic study on a substrain, the DBA/2NNia mouse. ² By 6 months of age, in the majority of these mice, peripheral anterior synechiae developed that were associated with iris atrophy and pigment dispersion. These changes were progressive and affected all mice by 9 months of age. At later ages (12–24 months), the retinal diseases associated with secondary glaucoma, such as retinal ganglion cell loss and optic nerve atrophy, became more prominent. ² A similar retinal disorder has recently been reported in another substrain, the DBA/2J mouse. ^{3 4}  

Genetic studies in these mice ⁵ have indicated that the initial disease of iris stroma atrophy and iris pigment dispersion occurs in inbred DBA/2J mice that are homozygous for two different alleles, one located on chromosome 4 and one on chromosome 6. In mice homozygous for only one of these alleles, a less severe disease develops. An important abnormality that develops in DBA/2J mice is the increase in intraocular pressure (IOP) beginning at 6 to 7 months and coinciding with the presence of peripheral anterior synechiae, iris atrophy, and pigment dispersion. ^{3 4} Thus, it seems most likely that these changes in the iris and the anterior chamber angle impede aqueous outflow and give rise to secondary glaucoma in DBA/2 mice. 

Our interest in rodent glaucoma models is to facilitate the development of drugs that will prevent or delay retinal ganglion cell loss. The present study was focused on the potential of electroretinogram (ERG) measurements as a noninvasive method to monitor retinal dysfunction in DBA/2NNia mice and on the correlation between ERG changes and morphometry of the neuronal retina as a function of age. 

Methods

All animals in this study were treated in accordance with the ARVO statement for the Use of Animals in Ophthalmic and Vision Research. 

Electroretinography

Animals.

The 15 DBA/2NNia (DBA) mice used in the electrophysiological measurements were inbred progeny of mice previously described by and obtained from Sheldon et al. ² As control subjects, a group of 16 C57BL/6J (C57) mice were observed over the same period (2–15 months of age). Mice were maintained in clear plastic cages and subjected to standard light cycles (12 hours light/12 hours dark) and were fed a standard rodent diet ad libitum (rodent chow 5001; Ralston Purina, St. Louis, MO). 

ERG Recording.

Scotopic ERGs were simultaneously recorded from both eyes of dark-adapted (12–14 hours) chloral hydrate–anesthetized mice (500 mg/kg intraperitoneal), using custom-made gold-wire corneal contact lens electrodes. ⁶ This is a monopolar electrode with a translucent plastic speculum and an embedded gold-wire loop. The corneas were anesthetized with 0.5% proparacaine drops and the pupils dilated by topical 1% atropine. The electrode was coated with 1% methylcellulose before making contact with the cornea. An indifferent silver-needle electrode was placed subcutaneously in the scalp, and grounding was by a saline-soaked, cotton-wick electrode placed in the mouth. 

A photostimulator (model PS-22; Grass, Quincy, MA) that delivers white-light flashes of 10-μsec duration was used. At the cornea, the irradiance of the unattenuated flash beam was approximately 236 lux/sec. When a range of neutral density filters was used (attenuation from 0.0 log unit to 3.0 log units in 0.6-log unit steps), light stimuli were presented in decreasing order of attenuation with a 5-minute interval between successive stimuli. Amplitudes of a- and b-waves were recorded and analyzed as a function of the log relative stimulus energy to determine the filter that best discriminated between normal and glaucomatous mice. Responses were differentially amplified (1–1000 Hz), averaged, and stored (System Interface Unit; LKC Technologies, Gaithersburg, MD). A Naka–Rushton type of analysis was performed by computer (Advanced Analysis software; LKC Tchnologies) to fit the data to a hyperbolic tanh function. The a-wave amplitudes were measured from the baseline to the peak of the negative a-wave (baseline to trough), whereas the b-wave amplitudes were measured from the trough of the a-wave to the peak of the positive b-wave. The implicit times of both were measured from the same points. The recordings obtained from this analysis were tabulated for each pair of rodent eyes. Scotopic ERGs of DBA mice and C57 mice were followed by recordings at 1-month intervals. Data from mice of 5 to 15 months of age are presented, because there were no significant differences in ERG between glaucomatous DBA and control C57 mice aged 6 months or less (all data not shown). 

Light Microscopy and Quantitative Morphology

Thirty DBA and 22 C57 mice, 1 to 14 months of age, were studied to determine the thickness of the retinal layers. The animals were anesthetized by perfusion of Ito’s solution (2.5% glutaraldehyde, 2.5% paraformaldehyde, and 0.01% picric acid in 0.1 mM cacodylate buffer [pH 7.2]). The eyes were enucleated and bisected along the ora serrata after they were rinsed in cacodylate buffer. The posterior eye segment was cut sagittally, and the halves were postfixed in 1% OsO₄, dehydrated, and embedded in Epon. Semithin sections were cut from each block by microtome (Ultracut OmU3; Reichert Jung, Vienna, Austria) and stained with toluidine blue. 

Sections through the posterior eye segment were defined as central when the plane passed through the optic nerve. Serial sections of each eye, 1- to 2-μm thick, were examined to ensure similar locations of measurement for all eyes. The total thickness of the neural retina and the thickness of the different retinal layers were quantitatively evaluated at the same location of the central retina in each eye. The measuring field was defined by a distance of 100 μm from the optic nerve head rim, and the single-thickness measurements (5–10 per eye) were obtained within the next 300 μm peripherally. Measurements of retinal thickness were performed by a computer (Quantimed 500; Leica, Bensheim, Germany) that was connected to a light microscope with a fixed camera. The measurements were performed by two independent observers (TN, ACM). The data points in the figures represent means of five measurements at different points in the measuring field. 

Statistical Analysis

All ERG data files were imported into computer software programs (SAS ver. 6.0; SAS, Cary, NC ⁷ ) for further statistical analysis. When comparing differences between scotopic ERG a- and b-wave amplitudes at different log unit attenuation filters, Student’s t-test was performed to establish the filter that best discriminated normal C57 mouse eyes from glaucomatous DBA mouse eyes at 10 months of age. 

To evaluate differences of ERG recordings between eyes of DBA mice and C57 mice monthly from 5 to 15 months of age, only one randomly chosen eye from each mouse was included in the sample to give n = 15 eyes studied (RANBIN function of SAS ver. 6.0 ⁷ ), and Student’s t-test was applied. Amplitudes of contralateral eyes of DBA mice were correlated to each other by means of linear regression analysis to test the hypothesis of equivalence of the disease in both eyes of individual mice. 

The thickness of the entire retina, the thickness of the different layers, the number of nuclear layers, and the number of ganglion cells were evaluated. Two time intervals were defined: up to the age of 6 months, when IOP is at normal levels ^{3 4} and when the animals in this study had relatively fewer areas in the circumference of the eye in which the outflow pathways were blocked by iris synechiae, and from 7 to 14 months of age, with presumed increased IOP and with synechiae occluding the outflow pathways in the entire circumference of the eye. Analyses of covariance within these intervals were performed for the different measurements, adjusted for the animals’ age. Significance tests were applied to comparisons between glaucomatous and control eyes within each of the intervals and also to a comparison of the first versus the second time interval, with reference to the observed differences between glaucomatous and control eyes. 

Contralateral eyes were not considered to be independent of each other. Consequently, the analyses were performed using data from only one randomly chosen eye of each mouse. The level of significance determined was two-sided in all statistical testing. Analyses were performed by computer (SPSS for Windows; SPSS, Chicago, IL). Nonlinear curve fitting was performed using the Lowess algorithm. ⁸  

Results

Electroretinographic Responses to Different Light Intensities

ERG responses (a- and b-wave amplitudes) to flashes of increasing intensity for both eyes in a group of C57 (n = 12) and a group of DBA (n = 16) mice at 10 months of age are shown in Figure 1 . The amplitude difference for both a- and b-wave was maximal when the stimulus intensity was attenuated by the −1.2- or −0.6-log unit filters. Therefore, all further comparative ERG recordings on additional groups of mice were performed using the −1.2-log unit filter. At this stimulus intensity, saturated a- and b-wave amplitudes were still obtained, and marked oscillatory potentials were also apparent. Representative ERG recordings from C57 and DBA mice at the ages of 5, 10, and 15 months are shown in Figure 2 . 

Age-Related Changes in Scotopic Electroretinography

The age-related changes from 5 months in recorded scotopic ERG a- and b-wave amplitudes in eyes of both the C57 mice and the DBA mice are illustrated in Figures 3A and 3B , respectively (n = 15 mice at each time point except 15 months, at which n = 13 DBA mice). The a-wave amplitudes from DBA mice showed the first significant reduction at 7 months (P = 0.0018) and became further attenuated over the next 8 months (P < 0.0001). Mean a-wave amplitude was 118.4 ± 8.6 μV at 5 months of age, and 10.1 ± 11.8 μV at 15 months of age. The b-wave amplitudes appeared to be unaffected until 8 months of age (P = 0.0158) and became further attenuated over the next 7 months (P < 0.0001). Mean b-wave amplitude was 234.5 ± 21.5 μV at 5 months of age and 111.7 ± 13.4 μV at 15 months of age. By 15 months of age, an a-wave of more than 5 μV was detectable in only 9 of the 15 DBA mice, but a reduced b-wave was still evident in all the mice. 

By contrast, in C57 mice, a- and b-wave responses did not decrease significantly with age to 15 months. There was a gradual decline in the mean b-wave response from 237.1 ± 22.9 μV at baseline (5 months of age) to 212.3 ± 17.9 μV at the end of the study, but the difference did not reach statistical significance. In these animals, the difference between the mean a-wave amplitudes of 121.1 ± 16.5μ V at baseline and 112.1 ± 17.0 μV at the end of the study were also not statistically significant. 

Although both groups of mice showed no statistically significant age-dependent differences in the a-wave amplitude implicit time (Fig. 4A ), the b-wave implicit time was gradually prolonged in DBA mice beginning at 7 to 8 months (P = 0.020), the same age at which the b-wave amplitude began to decline, when compared with C57 mice (Fig. 4B) . In DBA mice, mean b-wave implicit time was 30.5 ± 1.7 msec at baseline (5 months), and 47.3 ± 4.6 msec at 15 months of age (P < 0.0001). 

Correlations of ERG Amplitudes between Contralateral Eyes

To test whether there was consistency in the progression of ERG changes in individual animals and between the two eyes of DBA mice, the ERG amplitude data were correlated for both eyes of each mouse in the group at 5 months (n = 15) and again at 15 months of age (n = 13; Figs. 5 6 ). Statistical analysis revealed highly significant correlations between a-wave amplitudes of ipsilateral and contralateral eyes in DBA mice at 5 months of age (r ² = 0.9038; P < 0.0001, Fig. 5A ), and at 15 months of age (r ² = 0.9651; P < 0.0001, Fig. 5B ). The b-wave amplitudes correlated similarly in these mice at 5 months of age (r ² = 0.9576; P < 0.0001, Fig. 6A ) but showed somewhat more variability at 15 months of age (r ² = 0.5760; P < 0.001, Fig. 6B ). 

Additionally, in the DBA group the individual three mice with the smallest initial ERG a-wave (Fig. 5) and b-wave (Fig. 6) amplitudes (low responders), and the three individual mice with the largest amplitudes (high responders) at 5 months were again compared with the DBA group as a whole at 15 months of age, to determine consistency of the degree of change in response of individual mice over time (these subgroups of mice indicated in Figures 5 and 6 by square and triangle symbols, respectively). Mice designated as low or high responders at 5 months of age (Figs. 5A 6A) do not necessarily continue in the same responder group when retested at 15 months of age (Figs. 5B 6B) . 

Anterior Eye Segment Morphology

The morphology of the anterior eye segment in the DBA substrain mice of different age groups (data not shown) was very similar to that described previously by John et al. ³ In 1- or 2-month-old animals, the anterior chamber angle and Schlemm’s canal were still open, and no iris synechiae were present. However, by 4 months of age, there were interindividual differences, with iris synechiae present in some animals. In some parts of the circumference of the eye synechiae occluded the outflow pathway, whereas in other areas, the outflow pathways were still open. All animals between 7 and 14 months of age showed iris synechiae that occluded the outflow pathways in the entire circumference of the eye. The ciliary body in the 7- and 8-month-old animals showed some morphologic changes compared with C57 mice, but ciliary processes covered by a two-layered epithelium were still largely present. In the 10- to 14-month-old mice, the ciliary processes were severely degenerated. 

Morphologic Changes of the Neuronal Retina

Representative cross-sections of DBA mouse retinas at 1, 7, and 10 months are shown in Figure 7 . Thinning of the retinal layers and loss of cells is apparent in the 7-month-old retina and more pronounced in the 10-month-old retina, compared with the retina from a 1-month-old mouse. 

Total Thickness of the Retina.

In the central retina of C57 mice (Fig. 8A ) the total thickness of the neural retina increased to a small extent from an average of 220 μm (195–245 μm) in 2-month-old animals to 253 μm (251–256 μm) in 4-month-old animals and decreased to a small extent thereafter. In the glaucomatous DBA mice, the overall thinning of the retina was more pronounced with age than in the C57 mice (Fig. 8A) and showed no early increase in thickness. In 2-month-old DBA mice, the mean thickness was similar to that in control animals (mean, 220 μm; 201–230 μm). It then started to decrease in 4-month-old animals and reached a mean of 172 μm (160–180 μm) in 10-month-old animals. The retina remained thinner in 14-month-old mice (mean, 192 μm; 155–214 μm). The difference in total retinal thickness in C57 mice more than 6 months of age compared with that seen in DBA mice of the same age group was significant. Because of the increase in thickness of C57 retinas at 4 to 6 months, there was also a significant difference in retinal thickness in the DBA group of animals aged between 1 and 6 months. 

Inner Retina.

In the ganglion cell layer (GCL), the number of cells in C57 mice remained nearly constant in animals up to 12 months of age and decreased slightly in 14-month-old animals (Fig. 8B) . By comparison, in the GCL of the DBA mice the number of cells remained equivalent to those in C57 mice up to 5 months of age but decreased markedly in animals 7 months of age and more. This decrease continued up to the oldest (14-month-old) mice studied (Fig. 8B ; Table 1 ). 

A pronounced decrease in thickness was also seen in the inner plexiform layer (IPL) of DBA mice compared with the age-matched control animals. This difference in thickness was most pronounced in the 4- to 10-month-old mice (Fig. 8C) . The difference between mouse strains for both age groups (1–6 and 7–14 months) was statistically significant (Table 1) . 

The inner nuclear layer (INL) of the DBA mice showed high variability in number of cell layers at all ages. The C57 control animals consistently showed approximately five cell layers in all age groups, but in the DBA mice the number of layers varied between four and six in 1- to 5-month-old animals and between three and six in the older animals (Fig. 8D ; Table 1 ). 

Outer Retina.

In the outer plexiform layer (OPL), a decrease in thickness was first evident in animals at 7 months of age and continued until 14 months of age (Fig. 8E) . The difference in thickness between DBA mice and C57 control animals was significant in the 7- to 14-month age group, but not in the 1- to 6-month group (Table 1) . 

In the DBA mice, the number of cell layers of the outer nuclear layer (ONL) showed a higher variability than in the control animals, which had a relatively constant thickness with age. Whereas in the eyes of control animals, there were 9 to 10 cell layers in all age groups, eyes in 1- to 2-month-old DBA mice had from 8 to 12 layers of cells, and, in 14-month-old mice, the eyes had 7 to 10 layers. Overall, in DBA mice there was a decrease of cells with increasing age (Fig. 8F) , which was not the case for C57 control animals. The length of the rod inner (RIS) and outer segments (ROS) of DBA mice shortened with age. The difference between control animals and DBA mice became most prominent in animals more than 10 months of age (Figs. 8G 8H) . The difference between control animals and DBA mice in the 7- to 14-month age group was highly significant (Table 1) . There was, however, already a significant difference in the length of the inner and outer segments in the younger age group (Table 1) . 

Discussion

Our studies show for the first time a progression of both morphologic and functional changes in the neuronal retina of DBA mice, and that some of the changes have a comparable time course for the two parameters. Correspondence in physiological and morphologic measurements in this mouse glaucoma model indicates the value of noninvasive ERG measurements in documenting retinal pathophysiology in rodent glaucoma models. ⁹  

For the DBA mice, changes in a- and b-waves over time correlated highly for both eyes of individual animals. However, there were significant interanimal differences in ERG patterns. For example, the mice that were most severely affected at 5 months of age were not necessarily the most severely affected at 15 months of age. These interindividual differences in progression of the pathologic phenotype may have been due to variations in expression of the different alleles on chromosomes 4 or 6 that are known to be responsible for the anterior segment disease in DBA mice. ^{3 5}  

It is generally accepted that a decrease in the a-wave ERG amplitude is most likely related to changes in the outer retina, whereas decrease in b-wave amplitude and lengthening of the implicit time are related to changes in the INLs and synaptic connections. Previously, a significant loss of retinal ganglion cells was found in the DBA mouse substrain at 8 to 24 months of age. ² In the present study we demonstrate a complex time course of retinal changes in these mice compared with control C57 mice. In particular, early changes in the IPL and, to a lesser extent, shortening of rods became evident in DBA animals at 4 months of age. By this age, there were areas of the circumference of the eye showing iris synechiae in some mice. It is possible that mice in this age group have increased eye volume due to a higher outflow resistance that does not elevate IOP but instead induces stretching of the globe and inner retina. It is only after the onset of secondary angle-closure glaucoma associated with increased IOP at 6 to 7 months that structural changes occur in other retinal layers. These include substantial thinning of the OPL, loss of cells in the GCL, and some reduction in the number of photoreceptor cells in the ONL (Table 1) . These latter findings on retinal morphology appear to correlate in time with the changes in ERG parameters that become significant at 7 months and progress in older DBA mice. Some of the morphologic retinal changes, especially reduction in thickness of the plexiform layers, appear not to progress after 10 months of age. This may be accounted for by the observed degeneration of the ciliary processes reducing aqueous humor formation, with a decline in IOP from the preceding elevated levels. ⁴ In addition, some of the retinal thinning occurring during the period of 6 to 9 months of age due to progressive ocular enlargement and stretching of the retina may be arrested when the IOP decreases again. 

A decrease in thickness of the entire retina has also been described in Japanese quails with sex-linked albinism and glaucoma. ^{10 11} These birds show development of angle-closure with an increase in IOP between 4 and 6 months of age, due to iris attachment to the posterior cornea. The described pathologic changes in these quails show a striking similarity to the findings in the DBA mouse angle-closure glaucoma model documented in the present study. 

The specific causes of the changes in ERG parameters in DBA mice are not clear. However, because the changes occurred after approximately 4 months of age and progressed with the onset of anterior segment diseases (posterior iris synechiae occluding aqueous outflow), they appear to be due primarily to retinal damage, particularly loss in the nerve fiber ganglion cell and plexiform layers (Figs. 8B 8C 8E) , resulting from increased IOP. In humans, several earlier clinical studies have reported scotopic ERG changes in advanced glaucoma. ^{12 13 14} In a recent study of glaucomatous subjects, ¹⁵ amplitude reductions and peak-time prolongations of various components of scotopic and photopic ERGs under different stimulus conditions were found, suggesting widespread outer retinal dysfunction in human glaucoma. Oscillatory potentials recorded by flash ERG were as frequently reduced as the pattern ERG amplitudes were, ^{16 17} and other components of the flash ERG showed even more significant changes in glaucomatous eyes in another study. ¹⁸ A general reduction of ERG amplitude was also reported for ocular hypertensive and glaucomatous subjects. ¹⁹ The much more pronounced ERG changes seen in the DBA mouse may be due to genetic factors that cause a more rapid progression of retinal damage from untreated angle-closure glaucoma in a short-lived animal species with a rod-predominant–retina, compared with glaucoma in humans. 

Few measurements have been reported of changes in the thickness of retinal layers in glaucomatous eyes other than in the optic nerve fiber layer and GCL. In patients with primary open-angle glaucoma, Zeimer et al. ²⁰ reported a decrease in total retinal thickness up to 34% (the thickness of individual retinal layers was not measured in this study). In human secondary angle-closure glaucoma, thinning of the outer retinal layers was found, resulting from a decrease in INL thickness, ²¹ as was damage to and loss of photoreceptors. ^{22 23}  

To summarize our results, we found that the DBA/2NNia substrain angle-closure glaucoma mouse showed pathologic changes similar to those in other secondary glaucomas. Thus, even if the disease is more extensive than that found in human angle-closure glaucoma, DBA mice have potential as a model for investigation of posterior segment alterations in secondary angle-closure glaucoma. 

 Supported by an unrestricted research grant from Research to Prevent Blindness; by Grants Ba-94 and SFB539(BII.2) from the German Research Foundation; by the May and Samuel Rudin Family Foundation; by the Ernst and Berta Grimmke Foundation; and by Grants EY01867 and EY11649 from the National Institutes of Health.

 Submitted for publication April 17, 2000; revised December 15, 2000; accepted January 12, 2001.

 Commercial relationships policy: N.

 

Figure 1.

 

Scotopic ERG a- and b-wave amplitudes in eyes (n = 16) of 10-month-old DBA and C57 mice as a function of the light-stimulus intensity.

Figure 2.

 

Representative ERG recordings comparing C57 and DBA mice at 5, 10, and 15 months of age.

Figure 3.

 

Scotopic ERG a-wave (A) and b-wave (B) amplitudes in DBA and C57 mouse eyes as a function of age from 5 months (n= 15 at each time point to 14 months) to 15 months (n = 13).

Figure 4.

 

Scotopic ERG a-wave (A) and b-wave (B) implicit times in DBA and C57 mouse eyes as a function of age from 5 months (n = 15 at each time point to 14 months) to 15 months (n = 13).

Figure 5.

 

Correlation between the two eyes of individual DBA mice (○, ▴, ▪) for ERG a-wave amplitude at 5 months of age (n = 15; A) and at 15 months of age (n = 13; B). The three individual animals with the highest amplitudes (▴) and with the lowest amplitudes (▪) at 5 months of age are indicated.

Figure 6.

 

Correlation between the two eyes of individual DBA mice (○, ▴, ▪) for the ERG b-wave amplitude at 5 months of age (n = 15; A) and at 15 months of age (n = 13; B). The three individual animals with the highest (▴) and the lowest (▪) b-wave amplitudes at 5 months of age are indicated.

Figure 7.

 

Semithin sections through the central retina of 1-, 7-, and 10-month-old DBA glaucomatous mice (magnification, ×200). The thinning of the neural retinal layers and depletion of the neurons (as indicated in Table 1 ) in DBA retinas are shown by comparison of 7- and 10-month retinas with a 1-month-old retina. NFL, nerve fiber layer; GCL, ganglion cell layer; IPL, inner plexiform layer; INL, inner nuclear layer; OPL, outer plexiform layer; ONL, outer nuclear layer; RIS, rod inner segment layer; ROS, rod outer segment layer.

Figure 8.

 

Individual measurements and regression analyses of thickness or number of cells in retinal layers of DBA (•) and C57 mice (○) as a function of age from 1 to 14 months. (A) Total thickness, (B) GCL, (C) IPL, (D) INL, (E) OPL, (F) ONL, (G) RIS, and (H) ROS (abbreviations as in Fig. 7 ).

Table 1.

 

Comparison of Age-Dependent Changes of Layer Thickness in the Central Neural Retina of DBA Mice and C57 Control Mice

Layer of Retina	Young Animals (1–6 mo)			Old Animals (7–14 mo)
		P	F (df = 1.24)	P	F (df = 1.23)
GCL*	P = 0.332	F = 0.98	P < 0.001	F = 43.1
IPL	P < 0.001	F = 37.2	P < 0.001	F = 14.4
INL, †	P = 0.316	F = 1.05	P = 0.173	F = 1.98
OPL	P = 0.820	F = 0.05	P < 0.001	F = 83.3
ONL, †	P = 0.737	F = 0.12	P = 0.061	F = 3.87
RIS	P < 0.001	F = 13.1	P < 0.001	F = 40.1
ROS	P < 0.002	F = 12.5	P < 0.001	F = 18.5

 Analyses of covariance of the grouped DBA (n = 30) and C57 (n = 22) mice matched for equivalent age.

*  Based on cell count.

†  Based on number of cell layers.