Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Apr 10.
Published in final edited form as: Curr Eye Res. 2011 Feb 10;36(5):481–491. doi: 10.3109/02713683.2010.549600

Detection of Visual Deficits in Aging DBA/2J Mice by Two Behavioral Assays

Krsna Vedantha Rangarajan 1, Courtney Lawhn-Heath 1, Liang Feng 1, Ted S Kim 1, Jianhua Cang 1, Xiaorong Liu 1
PMCID: PMC13063309  NIHMSID: NIHMS2159381  PMID: 21309689

Abstract

Purpose:

The DBA/2J mice have been used as an animal model for human pigmentary glaucoma. However, these mice develop various degrees of disease symptoms at different ages, making it difficult to detect pathological changes of retinal degeneration at glaucoma onset. The purpose of this study is to develop a non-invasive assay to identify individual mice that develop visual deficits.

Materials and Methods:

We apply two behavioral tests, a swimming test of visual discrimination and a test of optomotor response, to identify glaucomatous DBA/2J mice. We then examine whether the elevation of intraocular pressure (IOP), the common risk factor for glaucoma, affects visual performances of the DBA/2J mice. We further compare the retinal ganglion cell death, one of the signature glaucoma symptoms, in mice with normal behavior with those with poor visual performances.

Results:

Our data demonstrate that (1) the onset of visual deficits in DBA/2J mice is around 7 months of age; (2) within each age group, there are various degrees of visual deficits; and (3) the percentage of mice exhibiting visual deficits increases with age and their visual capacities decrease gradually. Furthermore, the poor visual performances of DBA/2J mice do not correlate with the elevation of IOP. Importantly, compared to mice with normal visual performances in the same age group, mice with poor visual performances exhibit significant loss of retinal ganglion cells.

Conclusions:

Our studies establish a reliable behavioral assay to identify glaucomatous DBA/2J mice, thus making it possible to examine subtle pathological changes and molecular mechanisms in glaucoma pathogenesis with a relatively small number of samples.

Keywords: DBA/2J mice, Glaucoma, Intraocular pressure, Optomotor response test, Retinal ganglion cells, Visual water task

INTRODUCTION

Pigmentary glaucoma is a common form of secondary open angle glaucoma.1 In pigmentary glaucoma, pigment granules are released into the aqueous humor, leading to the blockage of aqueous humor drainage and intraocular pressure (IOP) elevation.1-3 These ocular abnormalities are followed by cupping of the optic nerve, optic nerve atrophy, and retinal ganglion cell (RGC) death.2,3 An animal model for human pigmentary glaucoma is the DBA/2J mice, which develop age-related progressive glaucoma initiated by iris atrophy and pigment dispersion.4-8 Even though the DBA/2J mice are inbred with a fixed genetic background, they exhibit a high degree of variability in developing glaucoma symptoms.8 It is estimated that the disease onset in DBA/2J mice is around 4–8 months of age, and the RGC loss and elevation in IOP become more evident around 9–13 months.5,6,8-10 At the age when most of the mice exhibit glaucoma symptoms, there is still a wide range of phenotypes from almost normal retina to severe RGC degeneration.8 Consequently, large cohorts of mice must be used in a single experiment to achieve statistical significance; and because of the individual differences observed within each age group, it is difficult to determine the temporal pattern of retinal degeneration and its underlying mechanisms during glaucoma pathogenesis.

The purpose of this study is to develop a non-invasive behavioral assay to identify individual mice that have developed visual deficits. Consequently, we could be able to detect pathological changes in the retina with fewer but more homogenous samples than classified based purely on age. Here we apply two established behavioral tests11,12 to examine changes of animal’s visual capacities (e.g., Refs. 13-16). Specifically, a swimming test of visual discrimination and a test of optomotor response are applied to the aging DBA/2J mice. Our data demonstrate that the visual capacities of DBA/2J mice gradually decrease with age, and within each age group, individuals exhibit various deficits in visual capacities. Importantly, the poor visual performances of the DBA/2J mice correlate with RGC loss, one of the most important glaucoma symptoms. Together, our studies provide reliable behavioral assays to identify individual glaucomatous mice with visual deficits. Mice can then be classified into different subgroups based on their visual performances, making it possible to detect pathological changes in the retina with a relatively small number of samples.

MATERIALS AND METHODS

Animals

DBA/2J and C57BL/6J mice were purchased from the Jackson Laboratory (Bar Harbor, Maine, USA) and raised at Northwestern University’s Animal Care Facility. All animals were used in accordance with protocols approved by Northwestern University Institutional Animal Care and Use Committee.

Behavioral Assessments of Visual Acuity

Two behavioral methods were used to assess mouse visual capacities—(1) the visual water task12 and (2) the optomotor response test11 (Cerebral Mechanics, Lethbridge, Canada).

Visual Water Task

The testing apparatus was a custom-made trapezoidalshaped pool filled with 22°C water to a depth of 15 cm (Figure 1A). Two monitors were placed side-by-side next to the transparent wide end wall. A vertical stationary low-frequency sinusoidal grating (0.12 cycle per degree, cpd), a standard psychophysical technique, was used.12,17 The black and white levels of the stimulus were approximately at 0.3 cd/m2 and 46 cd/m2, respectively. It was displayed on one of the monitors under which a removable platform was submerged ~1 cm below the water’s surface, while the other displayed uniform gray of the same mean luminance (~23 cd/m2). The visual stimulus was switched between the two monitors from trial to trial in a pseudo-random pattern. A midline divider was placed between the monitors to partially bisect the tank to create a choice line. Mice were released from the short end and swam toward the monitors. Once crossing the choice line, the animal was considered to have made a choice. The animal was rewarded quickly by being removed from the water if it chose the correct side; and it was forced to swim longer if it chose incorrectly. Each training session contained 10 trials and the total training sessions were clamped at eight. For aged animals, we offered four to six more training sessions (see an example in Figure 5A) to confirm their training performances. “Trainable” was defined as that the animal achieved three consecutive sessions of ≥ 80% correct (% of trials with the mouse swimming to the visual stimulus), and “Non-Trainable” was that the animal did not achieve three consecutive sessions of ≥ 80% correct even with extended training sessions. Only age-matched naive mice were compared in their training performances.

FIGURE 1.

FIGURE 1

Open in a new tab

Age-dependent decrease of visual performance in DBA/2J mice examined by a swimming test of visual discrimination. (A) Top view of the visual water task, in which the mouse was trained to associate a sinusoidal grating with escaping from the water. (B) Performances in training of different age groups. At the first training session, all DBA/2J mice swam randomly to either side (~50% of correct, dotted line). With increasing training sessions, young mice learned to swim towards to the visual stimulus, while most aged mice still swam randomly to both sides. (C) The visual performances of individual mice at the 8th training session (dashed line: 80% correct). Mice were defined as “trainable” when they achieved above 80% correct in more than 3 consecutive sessions. (D) The percentage of trainable DBA/2J mice progressively decreased with age. Error bar represents standard error of the mean (S.E.M., same in Figs. 2-5).

FIGURE 5.

FIGURE 5

Open in a new tab

Non-trainable DBA/2J mice exhibit a reduction in retinal ganglion cell density compared to age-matched trainable mice. (A) Representative examples of performances in the visual water task from trainable (labeled as T) and non-trainable (labeled as NT) mice. (B, D) Retinal sections (B) and whole-mounted retinas (D) from the same age groups shown in (A) were immuno-labeled with Brn-3a. Scale bars: 20 μm. (C) Quantifications of Brn-3a positive RGCs in retinal sections showed a significant decrease in RGC density in NT mouse-group compared to young and T groups. (E) Quantifications of Brn-3a positive RGCs in whole-mounted retinas also showed a decrease in RGC density in non-trainable mice. (F) Top: Orthogonal view of confocal images of calretinin-labeled amacrine cells in the whole-mounted retina. Bottom: calretinin-positive cells in the inner nuclear layer (INL) and the ganglion cell layer (GCL). (G) Calretinin-labeled cells in the GCL from 4m and 16m old mice. Scale bars: 20 μm. (H) The density of calretinin-positive cells in the GCL was not significantly different between trainable and non-trainable groups. *P < 0.05; ***P < 0.001 in the Student’s t-test.

Trainable mice were able to be further tested for acuity according to the established protocol.12,14 In brief, the spatial frequency of the testing grating was gradually increased starting from 0.12 cpd. For low frequencies (0.12–0.24 cpd), three to six trials were performed; for higher frequencies (0.24–0.6 cpd), blocks of ten trials were offered to the animal at each frequency. If the animal’s performance was ≥ 80% correct in three consecutive sessions, the spatial frequency was increased by adding one full cycle on the screen. This process continued until the success rate fell to chance (~ 50% correct). The last few values of performance percentage, starting from the last 100% correct, were fitted linearly, and the visual acuity was calculated as the spatial frequency at 70% correct from the fitted line.14 Typically, the training and the acuity test required 4–6 days with about 150 trials per animal in total.

After initial training and testing, if their acuities needed to be re-examined in the following months, the mice were given two to four “reminder” training sessions before testing. The acuity test was carried out when the animal had two consecutive sessions with ≥ 80% correct during the reminder training. If the animal’s performance was still unstable after four reminder sessions, two additional training sessions were performed. We classified the animal as “non-trainable” if it did not achieve two consecutive sessions of ≥ 80% correct even with six training sessions.

Optomotor Response Test

The subject mouse was placed and allowed to move freely on an elevated platform surrounded by four computer monitors with mean luminance of ~39 cd/m2 at the level of platform (Figure 3A). The monitors displayed horizontally drifting sinusoidal gratings as visual stimuli. The moving direction of the grating was alternated consecutively between clockwise and counterclockwise. The animal’s movement in-concert with the drifting gratings within 15 sec after stimulus onset was considered “positive.” The highest responseeliciting spatial frequency was defined as the animal’s visual acuity.11 The optomotor tests were performed two to four times for individual animals. Two trained observers carried out the test independently. When the acuities tested by the two observers were within 10% difference, we averaged the two readings as the final acuity; when the two acuities were more than 10% different, the two observers monitored the mouse’s visual behaviors together and reached an agreement on what was real head-tracking movement, and then re-tested the animal independently. In most cases, the difference of the two readings fell into the 10% range.

FIGURE 3.

FIGURE 3

Open in a new tab

Age-dependent decrease of visual acuity in DBA/2J mice measured by an optomotor response test. (A) Top view of the optomotor apparatus. The mouse was placed on a central platform and its spatial vision was determined via reflective head-tracking movements. (B) The visual acuity of DBA/2J mice decreased with age, while C57BL/6J mice maintained their acuity. *P < 0.05; **P < 0.01; ***P < 0.001 in One-way ANOVA post-Tukey test. (C) Comparison of visual acuities of 7-month-old C57BL/6J with DBA/2J mice. ***P < 0.001 in the Student’s t-test. (D) Trainable mice in the visual water task exhibited higher acuities in the optomotor response test compared to non-trainable mice. *P < 0.05 in the Student’s t-test. (E) Comparison of acuities measured by the two behavioral tests showed that acuity measured by the visual water task was generally higher than that in the optomotor test. The trend line was forced through 0 (Y = 1.19 X, and correlation coefficient R = 0.15).

In total, we trained 66 mice in the visual water task and examined 41 mice with the optomotor test. Thirty-five out of 66 mice (53%) were tested in the visual water task, and then examined by the optomotor test. Thirteen out of the 35 mice were trainable and 22 were non-trainable in the visual water task.

Histology

Retinas were dissected and fixed in 4% (w/v) paraformaldehyde in 0.1 M phosphate-buffered saline (PBS, pH 7.4) at different ages.18 Cryostat sections or whole-mount retinas were prepared.14 For sections, eyes were stored overnight in 20% sucrose in 0.1 M PBS at 4°C and were embedded in optimum cutting temperature compound (Tissue-Tek, Elkhart, Indiana, USA). Sections of the retina were cut perpendicular to the vitreal surface with a cryostat at 14–16 μm. To reduce variability, we only analyzed sections containing the optic nerve. The primary antibodies included anti-mouse Brn-3a (1:100, Millipore) and anti-calretinin (1:1000, Millipore).19 To visualize binding of the primary antibodies, sections or whole-mounted retinas were incubated in secondary antibody conjugated to Alexa Fluor 488, or Alexa Fluor 594 (diluted 1:500–1:1,000; Molecular Probes, Eugene, Oregon, USA). The incubation lasted 1–2 hr at room temperature for sections or overnight at 4°C for whole-mount retinas.

For retinal sections, images were captured with a Zeiss Observer A1 microscope; for whole-mount retinas, images were captured with a Zeiss Pascal confocal microscope or Observer A1 (Zeiss, Thornwood, New York, USA).20 Immuno-positive cells from six to ten fields of each retina were counted by two trained persons blind to the animal’s visual performances in behavioral assays.

IOP Measurement

The IOPs of both eyes were measured in adult mice using a rebound tonometer (TonoLab, Colonial Medical Supply, Franconia, NH).21,22 Six measurements of each eye were recorded. Because no significant difference was seen between the two eyes (data not shown), we averaged the IOPs of the two eyes for individual animals. Age-matched C57BL/6J mice were used as controls.

Statistical Analyses

A One-way ANOVA test was performed using GraphPad Prism (La Jolla, California, USA) to compare multiple samples, and Student’s t-tests were performed to compare paired samples. Correlation coefficient R was also calculated between two variables to measure their linear dependence.

RESULTS

DBA/2J Mice Exhibit Age-Dependent Decrease of Performance in Visual Water Task

We examined the visual performance of aging DBA/2J mice using a visual water task. In this test, mice were forced to discriminate between one monitor displaying a sinusoidal grating and the other of uniform gray (Figure 1A). Animals underwent eight sessions of training with each session consisting of ten individual trials. An individual trial was considered correct if the mouse swam directly towards the stimulus-displaying monitor. The visual performance was recorded as the percentage of correct trials.

We found that the DBA/2J mice’s capability to perform the visual water task gradually decreased with age (Figure 1B-D). At the 1st training session, all mice swam randomly to either side regardless of the display (Figure 1B). After eight training sessions, almost all 3–4-month-old and 6-month-old DBA/2J mice were able to perform the task (≥ 80% correct, Figure 1B-D). At the end of 8th training session, the visual performances of 3–4-month-old and 6-month-old mice were 96 ± 2% (n = 5), and 83 ± 3% (n = 10), respectively (Figure 1B). We plotted individual visual performances at the 8th training session and found that all of these mice had performances of ≥ 80% correct in this single session except one 6m old mouse (Figure 1C). We called mice “trainable” if they achieved above 80% correct in more than 3 consecutive sessions. By this criteria, almost all 3–4 month-old (5 out of 5) and 6-month-old (9 out of 10) mice were trainable (Figure 1D) and we further tested their visual acuities. The acuity of 4-month-old DBA/2J was 0.46 ± 0.08 cycle per degree (cpd), similar to C57BL/6J (0.53 ± 0.06, p = 0.8 in Student’s t-test, Figure 2A). Trainable 6-month-old DBA/2J mice also exhibited similar acuity (0.40 ± 0.02) as age-matched C57BL/6J mice (0.43 ± 0.02, p = 0.3, Figure 2A).

FIGURE 2.

FIGURE 2

Open in a new tab

Young DBA/2J mice have normal visual acuity and the decreasing of visual performance in aged DBA/2J mice is unlikely due to aging itself. (A) Trainable DBA/2J mice exhibited comparable acuity as age-matched C57BL/6J mice. (B) Comparison of the training performances between aged C57BL/6J and DBA/2J mice. *P < 0.05; **P < 0.01; ***P < 0.001 in Student’s t-test.

An obvious decrease in visual performance was observed by 7 months of age (Figure 1B-D). At the end of the training sessions, the visual performances of 7-month-old mice were 79 ± 7% (Figure 1B). Only 60% of 7-month-old mice were trainable (n = 11, Figure 1C, D) and these trainable mice exhibited normal acuity (0.41 ± 0.03) as age-matched C57BL/6J (p = 0.7, Figure 2A). The visual performances further decreased when the DBA/2J mice were one month older. At the end of the 8th training session, the visual performance of 8-month-old mice was 68 ± 5%, and only 41% of them were trainable (n = 17, Figure 1B-D). Almost all of the 9–10-month-old (8 out of 9) and 12–14-month-old (14 out of 14) mice became “non-trainable” as they still swam randomly towards to either side after eight training sessions (Figure 1B-C). Taken together, our data show that we could detect visual deficits in DBA/2J mice around 7 months of age by the visual water task; and that the percentage of mice that could perform the visual water task gradually decreases with age.

To confirm that the decrease of training performance is not due to aging itself, we compared the training performances between aged C57BL/6J and DBA/2J mice (Figure 2B). We offered four to six more sessions of extended training for 12–14-month-old mice to confirm whether they were able to perform the task (Figure 2B). As shown in Figure 2B, it did take more sessions for these aged C57BL/6J to “learn” the visual water task, but five out of six C57BL/6J mice became trainable after 14 training sessions. By contrast, the performances of aged DBA/2J mice were significantly worse than age-matched C57BL/6J mice. After 14 training sessions, none of the DBA/2J mice became trainable (n = 14, Figure 2B). Our data suggest that the decrease of visual performance in DBA/2J mice was not mainly affected by aging itself.

Visual Acuity Measured by Optomotor Test Gradually Decreases in Aging DBA/2J Mice

We tested the visual capacity of aging DBA/2J mice using another behavioral assay, the optomotor response test (Figure 3A). The optomotor test does not require training because it measures reflexive head movements.11 Consequently, it is less time-consuming than the visual water task. On the other hand, DBA/2J mice were more difficult to test because of their high activity level and the lack of attention paid to the visual stimulus compared to mice in other genetic backgrounds (personal observations). Consequently, we had two experienced observers performing the test independently, and repeatedly until we acquired reproducible measurements of their visual acuities (see Materials and Methods).

In the optomotor test, young DBA/2J mice exhibited normal acuity as age-matched C57BL/6J mice, similar to the results obtained in the visual water task. The mean acuity of 1–5-month-old DBA/2J mice (0.33 ± 0.04 cpd, n = 8) was similar to that of C57BL/6J mice (0.39 ± 0.01 cpd, n = 8, p = 0.24 in Student’s t-test, Figure 3B). By 7 months of age, the mean acuity of DBA/2J mice became significantly lower (0.23 ± 0.02 cpd, n = 10) than 1–5-month-old mice (p < 0.05 in One-way ANOVA Tukey’s Multiple Comparison Test, Figure 3B). Compared to age-matched C57BL/6J mice (0.38 ± 0.004 cpd, n = 5), the acuity of 7-month-old DBA/2J mice was also lower (p < 0.001 in Student’s t-test). Importantly, while the acuities of all C57BL/6J mice were within a narrow range of 0.35–0.4 cpd, individual DBA/2J mice exhibited various degrees of decrease in acuity (0.15–0.3 cpd, Figure 3C). Similarly, 9-month-old DBA/2J mice had much lower acuity (0.22 ± 0.02, n = 7) than age-matched C57BL/6J (0.41 ± 0.01, n = 10, p < 0.001 in Student’s t-test). At 12 or 14 months of age, most DBA/2J mice could not discern the visual stimulus (12 months: 0.15 ± 0.01 cpd, n = 10; 14 months: 0.06 ± 0.02 cpd, n = 6). By contrast, the acuities of C57BL/6J remained unchanged with age (12 months: 0.40 ± 0.01 cpd, n = 3; 14 months: 0.38 ± 0.01 cpd, n = 3; p > 0.05 in One-way ANOVA Test, Figure 3B). Together, our data show that visual acuities of DBA/2J mice progressively decrease with age and in each age group, individual mice exhibit different changes in acuity.

We compared two visual behavioral tests and found that they generally yielded consistent results (Figure 3D ,E). The trainable mice in the visual water task had higher acuities measured by the optomotor response test (0.27 ± 0.01, n = 13), compared to the non-trainable mice (0.14 ± 0.02, n = 22, p < 0.001 in Student’s t–test, Figure 3D). We further compared the acuities measured by the two tests (n = 13 mice) and found that the acuities were generally higher in the visual water task than in the optomotor test (Y = 1.19 X, and correlation coefficient R = 0.15, Figure 3E).

IOP Does Not Predict Visual Behavioral Performances

Elevated IOP is one of the common risk factors for glaucoma.5,6,23 Here we examined whether mice with high IOP had poor performance in visual behavior tests (Figure 4). In the visual water task, the average IOP for trainable mice was 20.9 ± 2.2 mmHg (n = 10), just below the value considered to be glaucoma-suspect in people (21 mmHg, red line; Figure 4A). Although, on average, non-trainable mice had slightly higher IOP (25.4 ± 2.9 mmHg, n = 8) than trainable mice (p = 0.22 in Student’s t-test), the DBA/2J mice exhibited a wide range of IOP regardless of their trainability (Figure 4A). Similarly, mice which could not detect the visual stimulus in the optomotor response test (recorded as 0 acuity) had IOP scattered between 10 and 40 mmHg (correlation coefficient R = −0.41, Figure 4B). Mice with high IOP (above 30 mmHg, dotted line in Figure 4B) also exhibited very different acuities in the optomotor response test (Figure 4B). Together, our data indicated that IOP elevation did not correlate with the poor performances in the behavior assays.

FIGURE 4.

FIGURE 4

Open in a new tab

Intraocular pressure (IOP) does not predict visual performances of DBA/2J mice. (A) No difference was detected in mean IOPs between the trainable and non-trainable DBA/2J mice (p = 0.2 in Student’s t-test). Red line: 21 mmHg, a value considered to be glaucoma-suspect in people. (B) No correlation was observed between IOP and the visual acuity measured by the optomotor response test (R = −0.41). (C) Four examples of changing IOP with age. When their IOP became elevated at different ages, mice #2 and #4 became non-trainable (labeled in red) and mouse #3 was still trainable in the visual water task. For mouse #1, its IOP fluctuated with age, but it became non-trainable after 8 months of age. (D) Mice #2 and #4 showed a good correlation between IOP elevation and decreased visual acuity, while mouse #3 had an elevated IOP but maintained relatively normal acuity.

We further followed the changes of IOP with age in four individual mice (Figure 4C, D). The IOP for mouse #2 became significantly higher after 8 months of age; correspondingly, it became non-trainable in the visual water task (Figure 4C). Although mouse #4 had elevated IOP at a much later age, it also became non-trainable in the water task (Figure 4C) and its visual acuity dropped sharply at the same time (Figure 4D). These two mice developed IOP elevation at different ages, but they exhibited a nice correlation of increased IOP and decreased visual performances. However, mouse #3 had elevated IOP after 8 months of age, though it had largely normal visual water training performance (Figure 4C) and normal acuity (Figure 4D). The IOP of mouse #1 fluctuated dramatically with age, but it became non-trainable after 8 months of age (Figure 4C). These results further suggest that the IOP elevation is not a good indicator for poor visual performances in DBA/2J mice.

RGC Loss Correlates with the Poor Performance in Visual Behavioral Assays

With the visual water task to identify individual mice with visual deficits, we could classify DBA/2J mice into trainable and non-trainable subgroups at the age of the glaucoma onset. We then tested whether the decrease of visual performance in the behavioral tests correlated with RGC loss, one of the most important glaucoma symptoms (Figure 5). Whole-mount or retinal sections were immuno-stained with the RGC marker Brn-3a antibody.18 Six- to 8-month-old mice were divided into trainable or non-trainable subgroups based on their performances in the visual water task (Figure 5A). As expected, trainable mice had normal RGC density (Sections: 8.2 ± 0.8/100 μm, n = 4; Whole-mount: 30.5 ± 2.1 RGCs/104 μm2, n = 8) as in young DBA/2J mice (Sections: 8.0 ± 0.5/100 um, n = 4, p = 0.83 in Student’s t-test; Whole-mount: 36.2 ± 2.5/104 um2, n = 4, p = 0.12, Figure 5B-E). By contrast, sectioned retinas from non-trainable 6- to 8-month-old mice had fewer RGCs (2.5 ± 0.1/100 μm, n = 5, p = 0.001); so did in the whole mount retinas (21.2 ± 4.6/104 μm2, n = 5, p = 0.03) than trainable mice (Figure 5B-E). As a control, 12- to 16-month-old DBA/2J mice exhibited severe RGC loss (Sections: 1.7 ± 0.6/100 μm, n = 4; p < 0.001; Whole-mount: 2.9 ± 0.7/104 μm2, n = 4, p < 0.001) than young ones (Figure 5B-E).

To examine whether amacrine cells, especially displaced amacrine cells in the ganglion cell layer (GCL), were affected in mice developing glaucoma, we immuno-stained the retinas with calretinin antibody, which mainly labeled a subset of amacrine cells in the inner nuclear layer (INL) and the GCL (Figure 5F). We measured the density of calretinin-positive cells in the GCL (Figure 5G, H). In trainable 6- to 8-month-old DBA/2J mice, whole mounted retinas contained 15.4 ± 0.9 calretinin-positive cells/104 um2 (n = 5), not significantly different from non-trainable mice (11.3 ± 1.5 calretin-in-positive cells/104 um2, n = 4; p = 0.06, Figure 5H). Twelve- to 16-month-old DBA/2J mice exhibited largely normal number of amacrine cells in the GCL (13.0 ± 0.9/104 μm2, n = 6), compared to young ones (14.9 ± 0.3/104 μm2, n = 4, p = 0.12, Figure 5F, G). Our data suggest that amacrine cell density in the GCL was not significantly affected in DBA/2J glaucomatous mice.

Taken together, our data suggest that mice exhibiting poor visual performances have significant RGC loss compared to the age-matched mice exhibiting normal visual performances. Our studies make it possible to detect pathological changes in the retina with fewer but more homogenous samples than classified based purely on age.

DISCUSSION

Our studies demonstrate that the visual water task and the optomotor response test could detect the visual deficits in DBA/2J glaucomatous mice. The visual water task measures the visual capacities of individual mice based on their performances in each trial (e.g., Ref. 13 and Figure 1); and the optomotor test provides a rapid assessment of visual acuity (e.g., Ref. 16 and Figure 2). It has been shown that DBA/2J mice had poor performance in the Morris water task.24 However, our data suggest no significant difference in the visual water performance between trainable DBA/2J mice and age-matched C57BL/6J mice (Figs. 1 and 2). Further experiments such as recording the swimming path and the total swimming time to the hidden platform may be needed to examine different aspects of the visual behaviors between DBA/2J and other lines, though our personal observations suggest there probably is no difference. Our data and observations are in agreement with studies that young DBA/2J mice do not have general impairments in motor coordination or learning.25,26 At the same time, we did notice that the optomotor reflex of DBA/2J mice was more difficult to test even before the onset of glaucoma due to their high activity level and the lack of attention to the visual stimulus. This characteristic of DBA/2J mice made it hard to distinguish reflexive visual responses from random movements, which could introduce more subjectivity to the optomotor test than that in many other lines such as C57BL/6J.

Different central visual structures are required to mediate the optokinetic head tracking and visual water swimming task in mice.14,17,27 The visual water swimming test is a behavioral test of cortical functions, and subcortical regions may be also needed.17 It is not clear which region controls the optokinetic head tracking, but the accessory optic system28 or the superior colliculus14 may be involved. Accordingly, degeneration in the higher visual centers may complicate interpretations of behavioral data in mice developing glaucoma.27,28 The trainable mice in the water task had relatively good visual acuity in the optomotor test, while non-trainable mice had poor acuity in the optomotor test (Figure 3D), suggesting that the two tests yielded generally consistent results in detecting visual deficits in aging DBA/2J mice. We noticed that the acuity measured by the visual water task was higher than that by the optomotor test (Figure 3E). It could be due to the different behavioral characteristic of DBA/2J mice in the optomotor response test as we discussed earlier, and/or that different central circuits were required.

Importantly, we could apply the two behavioral assays to screen for individuals with visual deficits not only in aging DBA/2J mice but also in other experimental glaucoma models. Based on the visual behavioral data, it is now possible to classify mice into different subgroups with different degrees of visual deficits. Furthermore, we could monitor the progress of glaucoma in individual mice since the behavioral assays allow an animal to be tested repeatedly (Figure 3C, D). This way we could reduce sample sizes in a single experiment and also classify samples into more homogeneous subgroups, thus making it possible to identify subtle pathological changes in the neural retina and to study the underlying molecular mechanisms at the onset of glaucoma.

Although elevated IOP is a causative risk factor, only half of glaucoma patients manifest elevated IOP.1,30 For some patients, treatment that lowers IOP does not help with the continuation of the visual field loss either.1,30 It seems that retinas are susceptible to elevated IOP but when and how that susceptibility triggers RGC degeneration and vision loss remain to be elucidated.3,31 For DBA/2J mice, IOP elevation occurs as a result of the iris disease.4-6 Typically, it peaks at around 9–10 months of age; however, even at this age, DBA/2J mice exhibit a wide range of IOP from below 10 to above 40 mmHg.5,23 Here our study shows that some mice with high IOP exhibit poor visual performance, but others do not (Figure 4), suggesting that like in human patients, IOP measurement itself is not a good indicator for visual behavioral performances of DBA/2J mice.

Future studies are needed to reveal how visual deficits at the behavioral level reflect the structural and functional degeneration in the neural retina, especially in the RGCs. In glaucoma patients, the visual field defects are usually accompanied with the decrease of RGC dendritic field size.32,33 A recent study shows that receptive field sizes increase in RGC axonal areas to compensate for the loss of RGC axons in glaucomatous eyes, resulting in the degradation of visual acuity and visual thresholds.34 Some studies suggest that the RGC loss is subtype-specific.35-37 For example, survived cells in DBA2/NNia mice have smaller cell sizes, implying preferential loss of larger RGCs.8,38 On the other hand, other studies indicate that in DBA/2J mice RGC death may be topological, but not cell-type-dependent.39 We may not expect a simple linear relationship between the RGC loss and the decrease of acuity because many factors could be involved at the visual behavioral level as we discussed above.

It is unclear whether there is a significant loss of amacrine cells in glaucomatous animals.40,41 For example, with IOP elevation, a reduction in the number of cholinergic amacrine cells but not dopaminergic amacrine cells was observed.40 Here our results suggest that calretinin-positive amacrine cells are not significantly changed in aging glaucomatous DBA/2J mice (Figure 5), consistent with previous finding that minimal change in amacrine cells was found in experimental glaucoma models.36

In summary, our studies indicate that the visual water task and optomotor response test detect visual deficits in glaucomatous mice and the poor visual performance reflects RGC loss. Our work provides the basis for future studies to examine subtle changes in retinal structure and function and their underlying molecular mechanisms in glaucoma.

ACKNOWLEDGMENTS

This work has been supported by a Midwest Eye-Banks Research Grant (to X.L.), and NIH-NEI grants R01EY018621 (to J.C.) and R01EY019034 (to X.L.).

Footnotes

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.

REFERENCES

  • 1.Farrar SM, Shields MB. Current concepts in pigmentary glaucoma. Surv Ophthalmol. 1993;37:233–252. [DOI] [PubMed] [Google Scholar]
  • 2.McKinnon SJ, Schlamp CL, Nickells RW. Mouse models of retinal ganglion cell death and glaucoma. Exp Eye Res. 2009;88:816–824. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Nickells RW. From ocular hypertension to ganglion cell death: A theoretical sequence of events leading to glaucoma. Can J Ophthalmol. 2007;42:278–287. [PubMed] [Google Scholar]
  • 4.Anderson MG. Mutations in genes encoding melanosomal proteins cause pigmentary glaucoma in DBA/2J mice. Nat Genet. 2002;30:81–85. [DOI] [PubMed] [Google Scholar]
  • 5.John SW, et al. Essential iris atrophy, pigment dispersion, and glaucoma in DBA/2J mice. Invest Ophthalmol Vis Sci. 1998;39:951–962. [PubMed] [Google Scholar]
  • 6.Chang B, et al. Interacting loci cause severe iris atrophy and glaucoma in DBA/2J mice. Nat Genet. 1999;21:405–409. [DOI] [PubMed] [Google Scholar]
  • 7.Libby RT, et al. Inherited glaucoma in DBA/2J mice: Pertinent disease features for studying the neurodegeneration. Vis Neurosci. 2005;22:637–648. [DOI] [PubMed] [Google Scholar]
  • 8.Schlamp CL, et al. Progressive ganglion cell loss and optic nerve degeneration in DBA/2J mice is variable and asymmetric. BMC Neurosci. 2006;7:66. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Danias J, et al. Quantitative analysis of retinal ganglion cell (RGC) loss in aging DBA/2NNia glaucomatous mice: Comparison with RGC loss in aging C57/BL6 mice. Invest Ophthalmol Vis Sci. 2003;44:5151–5162. [DOI] [PubMed] [Google Scholar]
  • 10.Reichstein D, et al. Apoptotic retinal ganglion cell death in the DBA/2 mouse model of glaucoma. Exp Eye Res. 2007;84:13–21. [DOI] [PubMed] [Google Scholar]
  • 11.Prusky GT, et al. Rapid quantification of adult and developing mouse spatial vision using a virtual optomotor system. Invest Ophthalmol Vis Sci. 2004;45:4611–4616. [DOI] [PubMed] [Google Scholar]
  • 12.Prusky GT, West PW, Douglas RM. Behavioral assessment of visual acuity in mice and rats. Vision Res. 2000;40:2201–2209. [DOI] [PubMed] [Google Scholar]
  • 13.Wong AA, Brown RE. Age-related changes in visual acuity, learning and memory in C57BL/6J and DBA/2J mice. Neurobiol Aging. 2007;28:1577–1593. [DOI] [PubMed] [Google Scholar]
  • 14.Wang L, et al. Direction-specific disruption of subcortical visual behavior and receptive fields in mice lacking the beta2 subunit of nicotinic acetylcholine receptor. J Neurosci. 2009;29:12909–12918. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Pinto LH, et al. Generation, characterization, and molecular cloning of the Noerg-1 mutation of rhodopsin in the mouse. Vis Neurosci. 2005;22:619–629. [DOI] [PubMed] [Google Scholar]
  • 16.Umino Y, Solessio E, Barlow RB. Speed, spatial, and temporal tuning of rod and cone vision in mouse. J Neurosci. 2008;28:189–198. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Prusky GT, Douglas RM. Characterization of mouse cortical spatial vision. Vis Res. 2004;44:3411–3418. [DOI] [PubMed] [Google Scholar]
  • 18.Liu X, et al. Brain-derived neurotrophic factor and TrkB modulate visual experience-dependent refinement of neuronal pathways in retina. J Neurosci. 2007;27:7256–6267. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Johnson J, et al. Vesicular glutamate transporter 1 is required for photoreceptor synaptic signaling but not for intrinsic visual functions. J Neurosci. 2007;27:7245–7255. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Liu X, et al. Regulation of neonatal development of retinal ganglion cell dendrites by neurotrophin-3 overexpression. J Comp Neurol. 2009;514:449–458. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Pease ME, Hammond JC, Quigley HA. Manometric calibration and comparison of TonoLab and TonoPen tonometers in rats with experimental glaucoma and in normal mice. J Glaucoma. 2006;15:512–519. [DOI] [PubMed] [Google Scholar]
  • 22.Nissirios N, et al. Noninvasive determination of intraocular pressure (IOP) in nonsedated mice of 5 different inbred strains. J Glaucoma. 2007;16:57–61. [DOI] [PubMed] [Google Scholar]
  • 23.Howell GR, et al. Absence of glaucoma in DBA/2J mice homozygous for wild-type versions of Gpnmb and Tyrp1. BMC Genet. 2007;8:45. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Crawley JN, et al. Behavioral phenotypes of inbred mouse strains: implications and recommendations for molecular studies. Psychopharmacol (Berl). 1997;132:107–124. [DOI] [PubMed] [Google Scholar]
  • 25.Nguyen PV, et al. Strain-dependent differences in LTP and hippocampus-dependent memory in inbred mice. Learn Mem. 2000;7:170–179. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Brown RE, Wong AA. The influence of visual ability on learning and memory performance in 13 strains of mice. Learn Mem. 2007;14:134–144. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Duncan RO, et al. Retinotopic organization of primary visual cortex in glaucoma: A method for comparing cortical function with damage to the optic disk. Invest Ophthalmol Vis Sci. 2007;48:733–744. [DOI] [PubMed] [Google Scholar]
  • 28.Simpson JI. The accessory optic system. Annu Rev Neurosci. 1984;7:13–41. [DOI] [PubMed] [Google Scholar]
  • 29.Crish SD, et al. Distal axonopathy with structural persistence in glaucomatous neurodegeneration. Proc Natl Acad Sci U S A. 2010;107:5196–5201. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Leske MC, et al. Factors for glaucoma progression and the effect of treatment: The early manifest glaucoma trial. Arch Ophthalmol. 2003;121:48–56. [DOI] [PubMed] [Google Scholar]
  • 31.Weinreb RN. IOP and the risk of progression to glaucoma. Graefes Arch Clin Exp Ophthalmol. 2005;243:511–512. [DOI] [PubMed] [Google Scholar]
  • 32.Nickells RW. Retinal ganglion cell death in glaucoma: The how, the why, and the maybe. J Glaucoma. 1996;5:345–356. [PubMed] [Google Scholar]
  • 33.Weber AJ, Harman CD. Structure-function relations of parasol cells in the normal and glaucomatous primate retina. Invest Ophthalmol Vis Sci. 2005;46:3197–3207. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Sharma SC. Changes of central visual receptive fields in experimental glaucoma. Prog Brain Res. 2008;173:479–491. [DOI] [PubMed] [Google Scholar]
  • 35.Quigley HA, Dunkelberger GR, Green WR. Chronic human glaucoma causing selectively greater loss of large optic nerve fibers. Ophthalmol. 1988;95:357–363. [DOI] [PubMed] [Google Scholar]
  • 36.Quigley HA, et al. Chronic glaucoma selectively damages large optic nerve fibers. Invest Ophthalmol Vis Sci. 1987;28:913–920. [PubMed] [Google Scholar]
  • 37.Shou T, et al. Differential dendritic shrinkage of alpha and beta retinal ganglion cells in cats with chronic glaucoma. Invest Ophthalmol Vis Sci. 2003;44:3005–3010. [DOI] [PubMed] [Google Scholar]
  • 38.Filippopoulos T, et al. Topographic and morphologic analyses of retinal ganglion cell loss in old DBA/2NNia mice. Invest Ophthalmol Vis Sci. 2006;47:1968–1974. [DOI] [PubMed] [Google Scholar]
  • 39.Jakobs TC, et al. Retinal ganglion cell degeneration is topological but not cell type specific in DBA/2J mice. J Cell Biol. 2005;171:313–325. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Hernandez M, et al. Immunohistochemical changes in rat retinas at various time periods of elevated intraocular pressure. Mol Vis. 2009;15:2696–2709. [PMC free article] [PubMed] [Google Scholar]
  • 41.Kielczewski JL, Pease ME, Quigley HA. The effect of experimental glaucoma and optic nerve transection on amacrine cells in the rat retina. Invest Ophthalmol Vis Sci. 2005;46:3188–3196. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES