July 2011
Volume 52, Issue 8
Free
Glaucoma  |   July 2011
Structural Correlation Between the Nerve Fiber Layer and Retinal Ganglion Cell Loss in Mice with Targeted Disruption of the Brn3b Gene
Author Affiliations & Notes
  • Andrew S. Camp
    From the Bascom Palmer Eye Institute, University of Miami Miller School of Medicine, Miami, Florida, and
  • Marco Ruggeri
    From the Bascom Palmer Eye Institute, University of Miami Miller School of Medicine, Miami, Florida, and
  • Gustavo C. Munguba
    From the Bascom Palmer Eye Institute, University of Miami Miller School of Medicine, Miami, Florida, and
  • Mary L. Tapia
    From the Bascom Palmer Eye Institute, University of Miami Miller School of Medicine, Miami, Florida, and
  • Simon W.M. John
    the Howard Hughes Medical Institute, The Jackson Laboratory, Bar Harbor, Maryland.
  • Sanjoy K. Bhattacharya
    From the Bascom Palmer Eye Institute, University of Miami Miller School of Medicine, Miami, Florida, and
  • Richard K. Lee
    From the Bascom Palmer Eye Institute, University of Miami Miller School of Medicine, Miami, Florida, and
  • Corresponding author: Richard K. Lee, Bascom Palmer Eye Institute, University of Miami Miller School of Medicine, 900 NW 17th Street, Miami, FL 33136; [email protected]
Investigative Ophthalmology & Visual Science July 2011, Vol.52, 5226-5232. doi:https://doi.org/10.1167/iovs.10-6307
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Andrew S. Camp, Marco Ruggeri, Gustavo C. Munguba, Mary L. Tapia, Simon W.M. John, Sanjoy K. Bhattacharya, Richard K. Lee; Structural Correlation Between the Nerve Fiber Layer and Retinal Ganglion Cell Loss in Mice with Targeted Disruption of the Brn3b Gene. Invest. Ophthalmol. Vis. Sci. 2011;52(8):5226-5232. https://doi.org/10.1167/iovs.10-6307.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose.: Mice with a targeted disruption of Brn3b (knockout Brn3b -/-) undergo the loss of a majority of retinal ganglion cells (RGCs) before birth. Spectral domain optical coherence tomography (SD-OCT) allows for the noninvasive examination of Brn3b -/- cellular loss in vivo.

Methods.: The central retinas of Brn3b -/- and phenotypically wild-type (Brn3b +/+ and Brn3b ±) mice were imaged by SD-OCT. The combined nerve fiber layer (NFL) and inner plexiform layer (IPL) were manually segmented and thickness maps were generated. The results were confirmed by histologic and immunofluorescence cell counts of the RGC layer (RGCL) of the same retinas.

Results.: The combined NFL and IPL of the Brn3b -/- retinas were significantly thinner, and the histologic cell counts significantly lower, than those of the phenotypically wild-type retinas (paired t-test; P < 0.01 and P < 0.01, respectively). The combined NFL and IPL thickness and the histologic cell count correlated highly (R 2 = 0.9612). Immunofluorescence staining revealed significant RGC-specific loss in Brn3b -/- retinas (paired t-test; P < 0.01). The distribution of combined central NFL and IPL loss was not localized or sectorial.

Conclusions.: The strong correlation between the combined layer thickness and histologic cell counts validates manual OCT segmentation as a method of monitoring cell loss in the RGCL. A retinal thickness map assessed if combined NFL and IPL thickness loss in Brn3b -/- eyes was topographically specific. Generalized RGC and combined NFL and IPL loss was observed in the Brn3b -/- retinas, in contrast to topographically specific RGC loss observed in glaucomatous DBA2/J eyes.

Animal models of glaucoma have significantly advanced research for this clinically important optic neuropathy. Glaucoma is a sight threatening optic neuropathy that is characterized by chronic progressive retinal ganglion cell (RGC) death. Animal models have allowed an understanding of the complex genetic and biochemical pathways involved in the death of RGCs. 1 Targeted disruption of the gene encoding BRN3B, a POU domain transcription factor, results in the loss of approximately 70% of RGCs prenatally in mice. 2 5 The RGC layer (RGCL) loss seen in Brn3b knockout mice (Brn3b -/-) is specific to RGCs and affects all RGC subtypes and dendritic projections to the inner plexiform layer (IPL). 6,7 BRN3B is crucial for early RGC differentiation, dendritic pathfinding, and central axonal projections. Brn3b knockout mice exhibit severe visual sensory defects secondary to RGC loss. 7 11  
Study of the role of BRN3B in RGC development and maintenance relies on the histologic examination of enucleated eyes. Spectral domain optical coherence tomography (SD-OCT) allows for the noninvasive visualization of retinal morphology and quantification of the nerve fiber layer (NFL) in vivo by measuring the intensity of backscattered light through retinal sections. 12 15 SD-OCT is currently used in the clinical assessment of glaucoma. Advances in SD-OCT technology have enabled researchers to study small-scale mouse eyes with progressively improved resolution. 16 19 Recent studies using manual segmentation of retinal layers illustrate the capability of SD-OCT to measure the thickness of retinal structures in mice but did not focus on topographical NFL degeneration across the entire retina or provide histologic correlation to actual cell counts. 20 22 Development of SD-OCT systems specifically designed for small animal imaging is expected to facilitate the use of OCT imaging for studies of retinal morphology and quantification through the assessment of retinal layer thickness in living mice. This will aid in noninvasive clinical follow-up for various experiments pertaining to retinal cell biology and retinal cell degeneration studies. 
We evaluated the capability of manual segmentation of SD-OCT images to identify differences in retinal layer thickness resulting from the loss of cells in the RGCL in living mice. We used three-dimensional (3D) manual segmentation of SD-OCT images to quantify and compare the combined NFL and IPL thicknesses in living Brn3b -/- and phenotypically wild-type (Brn3b +/+ and Brn3b +/−) mice noninvasively. We show that SD-OCT measurements of the combined NFL and IPL correlate well with histologic loss of cells in the RGCL. Immunofluorescence staining further detailed a significant loss specific to RGCs in Brn3b -/- retinas. In addition, we determined that RGC and combined NFL and IPL loss is diffuse in the Brn3b -/- central retina, unlike the focal loss of RGCs in the DBA2/J mouse glaucoma model. These preliminary findings set the stage for future studies using SD-OCT segmentation to track changes in retinal architecture secondary to cell loss. 
Methods
Animals
Experiments were performed in compliance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and within the guidelines of the University of Miami Institutional Animal Care and Use Committee. The mice used in this study were 2-month-old Brn3b +/+, Brn3b +/-, and Brn3b -/- mice (originally from Lin Gan, University of Rochester, NY) bred into a DBA/2J genetic background. Although DBA/2J develop increased intraocular pressure with age, no histologic changes have been reported in the first 3 to 5 months. 23 Brn3b +/+ and Brn3b -/- retinas are phenotypically wild-type and were used as controls. The genetic background of the mice used in this study was confirmed by PCR genotyping. In addition, the optic nerves from Brn3b -/- eyes were significantly smaller in caliber than Brn3b +/- and Brn3b +/+ nerves. Mice were kept on a 12-hour light/dark cycle and were fed standard rodent chow available ad libitum. 
SD-OCT Mouse Imaging
An SD-OCT system designed for small animal imaging (Bioptigen, Research Triangle Park, NC) was used for in vivo imaging of mouse retina. The axial resolution of the system is approximately 5 μm in retinal tissue. Mice were anesthetized with an intraperitoneal injection of a ketamine/xylazine mixture then placed on a positioning stage. The head was fixated on a stabilizing bar and a heating blanket was placed under the anesthetized mouse. The eyes were dilated with topical tropicamide and the corneas were kept moist during imaging with the regular application of artificial tears (Systane, Alcon, TX). Raster scans centered on the optic disc consisted of 1000 × 100 (horizontal × vertical) depth scans covering an area of 1 × 1 mm2 of the mouse retina. Alignment of the mouse took less than 1 minute, and image acquisition took approximately 2 seconds per eye. 
SD-OCT Retinal Layer Segmentation
Monitoring the thickness of the RGC containing NFL is of particular interest in studying mouse models of glaucoma. The IPL is also expected to be thinner in Brn3b -/- mice because of the loss of dendritic projections from the reduced numbers of RGCs in the RGCL and NFL. The NFL and IPL have similar scattering properties, and both appear as bright layers in the OCT images. For this reason, and because the NFL is on average very thin compared to the IPL, visual separation of the two layers is difficult, and therefore manual segmentation of the combined NFL and IPL was obtained for the OCT images of the Brn3b -/-, Brn3b +/-, and Brn3b +/+ mouse retinas. Thickness maps resulting from segmentation of the NFL and IPL boundaries were calculated together with the average combined layer thickness and SD for each eye. The retinal thicknesses of the Brn3b -/- and phenotypically wild-type mice were compared using a paired t-test. P < 0.05 was considered statistically significant. The genotype of two eyes was masked and segmentation was repeated with similar results to validate the reliability of this manual segmentation method. 
Histology and Cell Counts
Immediately after SD-OCT imaging, the eyes were harvested and fixed in formalin and were then processed and embedded in paraffin blocks. Five-micrometer hematoxylin and eosin (H&E)-stained sections across the entire retina of the globe were prepared. The retinal sections were examined by light microscopy using 400× total magnification. The total number of cells in the RGCL was manually counted across the entire retinal section in 80-μm increments. Cell counts of retinal sections taken from the peripheral retina were variable because of the preparation methods. The average was taken of the central 75% of the cell counts for each eye to minimize bias introduced by variable far peripheral retina cell counts. The central 75% of the cell counts of the Brn3b -/- and phenotypically wild-type mice were compared using a paired t-test. P < 0.05 was considered statistically significant. 
Immunofluorescence Staining
Five-micrometer sections of all globes were also prepared for immunofluorescence staining by fixation in formalin and 1X phosphate buffered saline (PBS; 127 mM NaCl, 2.7 mM KCl, and 10 mM phosphate) at 25°C. After fixation, globes were embedded in paraffin, sectioned, and mounted on glass slides. Sections were examined 160 μm to the left and right of the peak cell count as previously determined through histologic cell counts. Two sections per eye of the Brn3b -/-, Brn3b +/-, and Brn3b +/+ mice were stained. Primary antibodies were: rabbit anti-calretinin (1:100 dilution, #232A-74; Cell Marque, Rocklin, CA) and mouse anti-Brn3a (1:100 dilution, #SC-8429; Santa Cruz Biotechnology, Santa Cruz, CA). Secondary antibodies were: goat anti-rabbit Alexa 488 (1:100 dilution, #A11034; Molecular Probes, Eugene, OR) and goat anti-mouse Alexa 546 (1:100 dilution, #A11030; Molecular Probes). Sections were deparaffinized in xylene and sequential alcohol rinses, then treated with 95°C Trilogy (CMX833-C; Cell Marque) antigen retrieval reagent. Tissue sections were subsequently blocked with Rodent Block M (RBM96; BioCare Medical, Concord, CA) for 30 minutes to reduce nonspecific binding. Slides were incubated overnight at 4°C in PBS containing primary antibodies, washed with PBS, incubated in antibody buffer containing secondary antibodies for 1 hour at room temperature, washed with PBS, and mounted onto glass coverslips using DAPI vectashield (Vector, Burlingame, CA). Images were collected on an inverted microscope (Axiovert 200M; Carl Zeiss Meditec, Oberkochen, Germany) running the Zeiss AxioVision 4.7.2 on PC (Zeiss Inc., Thornwood, NY). Confocal microscopy images were obtained in a Leica TCP SP5 spectral confocal microscope (Leica, Exton, PA; 63×, water-immersion, 1.2 NA objective). Cell counts were taken across the extent of each retinal section examined. The total anti-Brn3a cell counts and anti-calretinin cell counts of the Brn3b -/- and phenotypically wild-type mice were compared using a paired t-test. P < 0.05 was considered statistically significant. 
Results
SD-OCT Imaging of Retinas of Mice with Targeted Brn3b Gene Disruption and Controls
SD-OCT images were obtained for eyes of 2-month-old mice: Brn3b -/-, Brn3b +/-, and Brn3b +/+. Brn3b +/- and Brn3b +/+ eyes are phenotypically wild-type. 4 All images were of sufficient resolution for use in this study despite moderate anterior chamber opacity related to transient cataract formation in some eyes. SD-OCT images were obtained near the center of the posterior pole from Brn3b +/+, Brn3b +/-, and Brn3b -/- retinas (Figs. 1A, 1C, 1E). The SD-OCT images were compared to histologic H&E-stained samples of the same eye (Figs. 1B, 1D, 1F). The anatomic layers of the retina can be seen in both the SD-OCT image and the H&E stain and are labeled accordingly (Fig. 1). A qualitative difference in both NFL density and cell count was observed between Brn3b -/- and phenotypically wild-type eyes. Phenotypically wild-type eyes have visually thicker RGCL cell counts than those of Brn3b deficient retinas. Histologic sections and SD-OCT images at various depths were compared and found to be in agreement with each other, demonstrating the anatomic correlation between histology and imaging (Fig. 1). 
Figure 1.
 
Representative comparison of SD-OCT B-scan retinal cross-sections (A, C, E) to hematoxylin and eosin (H&E)-stained retinal histologic sections (B, D, F) from the same mice. All retinal layers can be distinguished in both the SD-OCT images and H&E sections. Retinal layers are labeled as follows: neurofiber layer/retinal ganglion cell layer (NFL/RGCL), inner plexiform layer (IPL), inner nuclear layer (INL), outer plexiform layer (OPL), outer nuclear layer (ONL), photoreceptor inner segment/outer segment (IS/OS), and RPE. The NFL of the Brn3b -/- retina (A and B) is qualitatively thinner than that of the Brn3b +/+ (E and F) and Brn3b +/- (C and D) retinas.
Figure 1.
 
Representative comparison of SD-OCT B-scan retinal cross-sections (A, C, E) to hematoxylin and eosin (H&E)-stained retinal histologic sections (B, D, F) from the same mice. All retinal layers can be distinguished in both the SD-OCT images and H&E sections. Retinal layers are labeled as follows: neurofiber layer/retinal ganglion cell layer (NFL/RGCL), inner plexiform layer (IPL), inner nuclear layer (INL), outer plexiform layer (OPL), outer nuclear layer (ONL), photoreceptor inner segment/outer segment (IS/OS), and RPE. The NFL of the Brn3b -/- retina (A and B) is qualitatively thinner than that of the Brn3b +/+ (E and F) and Brn3b +/- (C and D) retinas.
Quantitative Analysis of Retinal Layer Thickness
OCT images were manually segmented for all eyes and thickness maps of the combined NFL and IPL were generated (Figs. 2A, 2B, 2C). The thickness map is displayed over a circular area with an equivalent diameter of 0.7 mm. Segmentation of the combined NFL and IPL was not accurate in the optic disc region because of the progressive taper of the retinal layers toward the optic nerve head. For this reason, thickness maps were calculated after removing data from a circular area with a 245-μm diameter centered on the optic disc. Color bars ranging from 0 to 100 μm are shown together with the thickness maps. 
Figure 2.
 
The representative three-dimensional segmentation of the NFL and IPL of Brn3b -/- (A), Brn3b +/- (B), and Brn3b +/+ (C) retinas yields a retinal thickness map. Thickness loss is uniform across the Brn3b -/- retina. Retinal artery locations in the OCT generated fundus images of the Brn3b -/- (D), Brn3b +/- (E), and Brn3b +/+ (F) mice match well with the retinal arteries seen in the thickness map.
Figure 2.
 
The representative three-dimensional segmentation of the NFL and IPL of Brn3b -/- (A), Brn3b +/- (B), and Brn3b +/+ (C) retinas yields a retinal thickness map. Thickness loss is uniform across the Brn3b -/- retina. Retinal artery locations in the OCT generated fundus images of the Brn3b -/- (D), Brn3b +/- (E), and Brn3b +/+ (F) mice match well with the retinal arteries seen in the thickness map.
The retinal blood vessel pattern can be distinguished on the thickness map as a thickening of the segmented complex layer. The locations of the retinal arteries correlate well with the arteries visualized in the raw SD-OCT Z-stack en face fundus images displayed in grayscale in Figure 2 (D- F). This validates the ability of the segmentation program to accurately track the contours of the retinal layers. The retinal thickness maps of the phenotypically wild-type retinas (Figs. 2B, 2C) are yellow-shifted when compared to those of the Brn3b knockout retina (Fig. 2A). This indicates that the phenotypically wild-type retinas have a greater combined NFL and IPL thickness than the knockout retina. The uniform color of the thickness map of the Brn3b knockout retina indicates a relatively even loss of combined layer thickness across the central retina. 
The average thickness and SD of the combined NFL and IPL was calculated for each eye using a previously described segmentation algorithm modified for use with images obtained from the Bioptigen system. 24 Brn3b -/- eyes have thinner inner retinal layers with no overlap of SDs compared to thicker, phenotypically wild-type mouse retinal layers (Fig. 3). The average thickness of the combined NFL and IPL of Brn3b -/- retinas is significantly less than that of phenotypically wild-type mice (paired t-test; P < 0.01). Phenotypically wild-type mice, Brn3b +/- and Brn3b +/+, have retinal layers of similar thickness, suggesting no Brn3b gene expression dosage effect on the combined NFL and IPL thickness (Fig. 3). 
Figure 3.
 
Retinal thicknesses and SDs (error bars) were calculated using data from the retinal thickness map. Brn3b -/- mice have thinner retinas and nonoverlapping SDs with phenotypically wild-type (Brn3b +/+ and Brn3b +/-) mice.
Figure 3.
 
Retinal thicknesses and SDs (error bars) were calculated using data from the retinal thickness map. Brn3b -/- mice have thinner retinas and nonoverlapping SDs with phenotypically wild-type (Brn3b +/+ and Brn3b +/-) mice.
Cell Counts and Retinal Layer Thickness Correlations
A comparison of RGCL cell counts across the entire retinas of H&E-stained Brn3b +/+, Brn3b +/-, and Brn3b -/- eyes shows a clear reduction in cell numbers across the entire retina of Brn3b -/- eyes. The cell counts of the Brn3b -/- retinas are consistently lower than the phenotypically wild-type Brn3b +/- and wild-type Brn3b +/+ retinas across the breadth of the eye (Fig. 4A). The averages of the top 75% of the H&E-stained RGCL cell counts for all eyes, with error bars representing the single maximal cell count per section of each eye, are shown in Figure 4B. The Brn3b -/- eyes have significantly lower RGCL cell counts than the phenotypically wild-type eyes (paired t-test, P < 0.01). 
Figure 4.
 
(A) Cell counts in the NFL of a Brn3b -/-, Brn3b +/-, and a Brn3b +/+ mouse shows a reduction in cell number in the Brn3b -/- NFL. (B) Averaging the top 75% of the cell counts for all mice shows a reduction in cell numbers in Brn3b -/- versus phenotypically wild-type retinas. Error bars represent the maximum cell count for each retina.
Figure 4.
 
(A) Cell counts in the NFL of a Brn3b -/-, Brn3b +/-, and a Brn3b +/+ mouse shows a reduction in cell number in the Brn3b -/- NFL. (B) Averaging the top 75% of the cell counts for all mice shows a reduction in cell numbers in Brn3b -/- versus phenotypically wild-type retinas. Error bars represent the maximum cell count for each retina.
The reduced cell count in the RGCL of Brn3b -/- retinas was confirmed and further explored through immunofluorescence staining. The cell counts in the RGCL of Brn3b -/- retinas are reduced in both the Brn3a-stained (Fig. 5B) and calretinin-stained (Fig. 5C) images when compared to the equivalently stained Brn3b +/- retinas (Figs. 5E, 5F, respectively) and Brn3b +/+ retinas (Figs. 5H, 5I, respectively). A significant reduction of both Brn3a- and calretinin-stained cells in the RGCL of Brn3b -/- retinas was observed when compared to phenotypically wild-type Brn3b +/- and Brn3b +/+ retinas (paired t-tests; P < 0.01). However, when the Brn3a-stained cell count was subtracted from the calretinin-stained cell count, no significant difference was found between Brn3b -/- retinas and phenotypically wild-type retinas (paired t-test; P = 0.34). The calculated average top 75% H&E-stained cell counts were plotted as an independent variable to the dependent SD-OCT calculated NFL and IPL thickness (Fig. 6). A correlation between the average cell count in the RGCL and the calculated thickness of the NFL and IPL was observed (R 2 = 0.9612; Fig. 6). 
Figure 5.
 
Representative comparison of flattened Z-stack images of Dapi- (A, B, C), Brn3a- (D, E, F), and calretinin-stained (G, H, I) retinal cross sections from Brn3b -/-, Brn3b +/-, and Brn3b +/+ mice. Retinal layers are labeled as previously indicated. The RGCL of Brn3b -/- retinas (A, D, G) has significantly fewer Brn3a- and calretinin-stained cells than the RGCL of Brn3b +/- (B, E, H) and Brn3b +/+ (C, F, I) retinas.
Figure 5.
 
Representative comparison of flattened Z-stack images of Dapi- (A, B, C), Brn3a- (D, E, F), and calretinin-stained (G, H, I) retinal cross sections from Brn3b -/-, Brn3b +/-, and Brn3b +/+ mice. Retinal layers are labeled as previously indicated. The RGCL of Brn3b -/- retinas (A, D, G) has significantly fewer Brn3a- and calretinin-stained cells than the RGCL of Brn3b +/- (B, E, H) and Brn3b +/+ (C, F, I) retinas.
Figure 6.
 
Scatter plot using NFL cell counts as the independent variable and calculated retinal thickness as the dependent variable. A strong correlation between cell counts and calculated retinal thickness is observed (R 2 = 0.9612).
Figure 6.
 
Scatter plot using NFL cell counts as the independent variable and calculated retinal thickness as the dependent variable. A strong correlation between cell counts and calculated retinal thickness is observed (R 2 = 0.9612).
Discussion
Study of the RGCL and the NFL has typically been dependent on the histopathologic analysis of retinas from enucleated eyes or retinal cell cultures. These techniques allow only a static picture of in vivo retinal changes. Most Brn3b-dependent RGCs undergo apoptosis perinatally or in early postnatal development, so Brn3b -/- retinas likely undergo a majority of their alterations before adulthood. 5 However, RGCs continue to express Brn3b throughout the lifetime of the cells, suggesting an important role for BRN3B in the maintenance of RGC survival. 25 The development of a Brn3b-dependent Cre/lox system for genetic ablation of adult Brn3b-expressing RGCs lays a framework for study of Brn3b expression and the role of Brn3b-expressing RGCs in adult mice. 26 As such, the development of a noninvasive retinal monitoring system capable of following individual mice across multiple time points may play an important role in the in vivo study of Brn3b in adult mouse RGCs. More importantly and generally, noninvasive imaging of the retina will allow for sequential in vivo study of the natural history of the changes the retina undergoes under certain model disease conditions, such as glaucoma and retinal degeneration, as is observed in human eyes. 
The SD-OCT imaging system (Bioptigen) used in this study provides high-resolution images of the retinal layers in living mice. OCT studies in human eyes have long shown the relationship between NFL thickness and optic nerve function in glaucomatous patients. 27,28 The small pupil size and optics of the mouse eye has slowed the development of an OCT system with similar resolution for mouse eyes. Now that such an animal SD-OCT imaging system is commercially available, research using mouse models can be expected to accelerate on a much greater scale. The results of our preliminary work show the utility of manual segmentation of SD-OCT images in determining retinal layer thickness as correlated to histologic cell counts. Ideally, automated segmentation will be forthcoming for this technology. We found a strong correlation between the histologic number of cells in the RGCL and the combined thicknesses of the NFL and IPL measured by SD-OCT (R 2 = 0.9612; Fig. 6). This finding supports observing changes in NFL and IPL thickness by SD-OCT longitudinally in vivo in individual mice as measures of RGCL loss associated with mouse models of glaucoma. This methodology could be used in other progressive mouse models of glaucoma to characterize the temporal relationship between thinning of the NFL and cellular loss in the RGCL by examining multiple timepoints in the progression of the disease. 
Our RGCL cell counts confirm previous histologic work showing RGC loss in Brn3b -/- mice. We show a significant reduction in the total H&E-stained RGCL cell count of Brn3b -/- retinas compared with phenotypically wild-type retinas (paired t-test; P < 0.01). The total cell loss in the RGCL is on a similar scale as the 30% previously reported. 4 Immunofluorescence staining and cell counts further characterize the cell types found in the RGCL of the mice used in this study (Fig. 5). Displaced amacrine cells are present in the RGCL of Brn3b -/- retinas in addition to RGCs, so it is necessary to determine that the total cell loss observed is not related to the loss of displaced amacrine cells alone. 29 Brn3a is a well characterized marker of RGCs, and calretinin stains both displaced amacrine cells and a majority of RGCs. 30 32 We show a significant loss of Brn3a-labeled RGCs in the Brn3b -/- retinas similar to the 70% loss of RGCs previously reported in Brn3b -/- mice 2,4 (paired t-test; P < 0.01). 
A significant loss in calretinin-labeled cells in the RGCL of Brn3b -/- retinas further confirms the total cell loss observed in the H&E-stained sections (paired t-test; P < 0.01). The lack of statistically significant cell loss after the subtraction of Brn3a-stained RGCs from the total calretinin-stained cell count suggests that the loss of RGCs is largely responsible for the total cell loss observed in Brn3b -/- retinas (paired t-test; P < 0.34). However, the lack of significance in our work could be related to sample size or the limitations of immunofluorescence staining. Future work could more accurately characterize cell types through retrograde labeling. 33 Displaced amacrine cells could also be tracked more accurately and specifically as better markers are developed. Additional work will be necessary to characterize the specific cell types lost in other glaucoma models. Despite these limitations, we believe that this work supports the use of SD-OCT measurements of RGCL thickness as a general marker of total cell loss in the RGCL and RGC-specific cell loss in the Brn3b -/- mouse model. 
Brn3b -/- mice have retinas that are 20% thinner than phenotypically wild-type mice. 4 Although all subtypes of RGCs are equally affected in Brn3b -/- mice, some subtypes are lost in an asymmetric topographic pattern. 6 More extensive topographic analysis of cell loss has been performed in other mouse models of glaucoma, such as the DBA2/J mouse. DBA2/J mice develop pigmentary glaucoma because of a mutation in genes encoding melanosomal proteins. 23,34,35 Analysis of total cell loss in retinas of aged DBA2/J mice shows damage to RGC axons beginning around the lamina, followed by clustered and fan-shaped areas of cell death across the retina with differential gene expression in areas of RGC loss. 36 40 A similar pattern of sectorial RGC loss has also been shown in a rat ocular hypertension model. 33 Our 3D thickness model of the NFL and IPL of Brn3b -/- eyes shows a symmetrical and nonlocalized loss of retinal thickness in the central retina (Fig. 2). These images suggest that, as a whole, cell loss in the NFL and IPL of Brn3b -/- mice follows a generalized pattern of RGC loss and is not regionally specific—a pattern of cell loss similar to that seen in the C57BL/6 mouse. 41  
Only one of the two RGC subtypes described by Lin et al. 42 showed any asymmetry in the Brn3b -/- model, and more than 10 subtypes exist, so it is possible that the asymmetry described is subtype-specific and overshadowed by losses among the rest of the subtypes. One shortcoming of this method of retinal segmentation is the inability of SD-OCT to distinguish between RGC subtypes, so localized changes in subtypes such as that described in aged DBA/2NNia mice will be missed. 43 The restriction of the OCT field to a 1-mm2 field surrounding the optic nerve head also may have missed topographical loss specific to the peripheral retina. Although analyses of topographical loss in other mouse models have shown a pattern of loss extending to the optic nerve head, a wider field will be necessary to show true homogenous loss. 36,37,40  
Other modalities, such as scanning laser polarimetry (SLP) and confocal scanning laser ophthalmoscopy (cSLO), can be used to image the NFL in vivo. However, like SD-OCT, these imaging modalities are unable to differentiate between RGC subtypes and are further hampered by their reliance on indirect measurements of NFL thickness. 15,44,45 The use of gene-based reporters allows for the examination of specific subtypes of RGCs to overcome the limitations in cellular subtype differentiation. GFP-labeling of RGCs in transgenic mice has been a particularly active field and provides a new method for researchers to distinguish RGC subtypes. 46,47 However, imaging of GFP-labeled retinal cells in vivo is limited by shallow image penetration and low image resolution in vivo. 48 Ongoing research in the field of molecular contrast OCT (MCOCT) works to combine the molecular specificity of fluorescent labeling with the greater penetration and higher resolution of SD-OCT images, 49,50 As this nascent field advances, there is great potential for contribution to the study of retinal changes in models of ocular disease. 
Our work illustrates the use of SD-OCT in the study of RGC loss and supports additional work to evaluate RGC loss in mouse models of glaucoma. The correlation between total RGCL cell counts measured histologically and SD-OCT measured combined NFL and IPL thicknesses validates segmentation of SD-OCT retinal layers as a marker for RGCL cell loss in vivo. Immunofluorescence work confirms that the difference in cell count is largely related to changes in total RGC counts in the Brn3b -/- retina. The 3D thickness map generated can be used to determine whether any observed cell loss occurs in topographically localized areas. These in vivo imaging methods will become more important as SD-OCT becomes widely available for the study of animal models for ocular disease, such as glaucoma, retinal degenerations, and other eye diseases with chronic progressive changes. 
Footnotes
 Supported in part by National Eye Institute Grants EY016775 (RKL), EY16112 (SKB), and EY011721 (SWMJ); a Research to Prevent Blindness Career Development Award (RKL); the Alpha Omega Alpha Carolyn L. Kuckein Student Research Fellowship (ASC); and the Fight for Sight Summer Student Fellowship (ASC). The Bascom Palmer Eye Institute is supported by a National Institutes of Health Core Grant NIHP301430 and an unrestricted grant from Research to Prevent Blindness. SWMJ is an Investigator of the Howard Hughes Medical Institute.
Footnotes
 Disclosures: A.S. Camp, None; M. Ruggeri, None; G.C. Munguba, None; M.L. Tapia, None; S.W.M. John, None; S.K. Bhattacharya, None; R.K. Lee, None.
References
McKinnon SJ Schlamp CL Nickells RW . Mouse models of retinal ganglion cell death and glaucoma. Exp Eye Res. 2009;88:816–824. [CrossRef] [PubMed]
Erkman L McEvilly RJ Luo L . Role of transcription factors Brn-3.1 and Brn-3.2 in auditory and visual system development. Nature. 1996;381:603–606. [CrossRef] [PubMed]
Gan L Wang SW Huang Z Klein WH . POU domain factor Brn-3b is essential for retinal ganglion cell differentiation and survival but not for initial cell fate specification. Dev Biol. 1999;210:469–480. [CrossRef] [PubMed]
Gan L Xiang M Zhou L Wagner DS Klein WH Nathans J . POU domain factor Brn-3b is required for the development of a large set of retinal ganglion cells. Proc Natl Acad Sci U S A. 1996;93:3920–3925. [CrossRef] [PubMed]
Xiang M . Requirement for Brn-3b in early differentiation of postmitotic retinal ganglion cell precursors. Dev Biol. 1998;197:155–169. [CrossRef] [PubMed]
Lin B Wang SW Masland RH . Retinal ganglion cell type, size, and spacing can be specified independent of homotypic dendritic contacts. Neuron. 2004;43:475–485. [CrossRef] [PubMed]
Badea TC Cahill H Ecker J Hattar S Nathans J . Distinct roles of transcription factors brn3a and brn3b in controlling the development, morphology, and function of retinal ganglion cells. Neuron. 2009;61:852–864. [CrossRef] [PubMed]
Erkman L Yates PA McLaughlin T . A POU domain transcription factor-dependent program regulates axon pathfinding in the vertebrate visual system. Neuron. 2000;28:779–792. [CrossRef] [PubMed]
Mu X Beremand PD Zhao S . Discrete gene sets depend on POU domain transcription factor Brn3b/Brn-3.2/POU4f2 for their expression in the mouse embryonic retina. Development. 2004;131:1197–1210. [CrossRef] [PubMed]
Qiu F Jiang H Xiang M . A comprehensive negative regulatory program controlled by Brn3b to ensure ganglion cell specification from multipotential retinal precursors. J Neurosci. 2008;28:3392–3403. [CrossRef] [PubMed]
Wang SW Gan L Martin SE Klein WH . Abnormal polarization and axon outgrowth in retinal ganglion cells lacking the POU-domain transcription factor Brn-3b. Mol Cell Neurosci. 2000;16:141–156. [CrossRef] [PubMed]
Baumal CR . Clinical applications of optical coherence tomography. Curr Opin Ophthalmol. 1999;10:182–188. [CrossRef] [PubMed]
Nassif N Cense B Park B . In vivo high-resolution video-rate spectral-domain optical coherence tomography of the human retina and optic nerve. Opt Express. 2004;12:367–376. [CrossRef] [PubMed]
Sakata LM Deleon-Ortega J Sakata V Girkin CA . Optical coherence tomography of the retina and optic nerve—a review. Clin Exp Ophthalmol. 2009;37:90–99. [CrossRef]
Townsend KA Wollstein G Schuman JS . Imaging of the retinal nerve fibre layer for glaucoma. Br J Ophthalmol. 2009;93:139–143. [CrossRef] [PubMed]
Cebulla CM Ruggeri M Murray TG Feuer WJ Hernandez E . Spectral domain optical coherence tomography in a murine retinal detachment model. Exp Eye Res. 2010;90:521–527. [CrossRef] [PubMed]
Li Q Timmers AM Hunter K . Noninvasive imaging by optical coherence tomography to monitor retinal degeneration in the mouse. Invest Ophthalmol Vis Sci. 2001;42:2981–2989. [PubMed]
Srinivasan VJ Ko TH Wojtkowski M . Noninvasive volumetric imaging and morphometry of the rodent retina with high-speed, ultrahigh-resolution optical coherence tomography. Invest Ophthalmol Vis Sci. 2006;47:5522–5528. [CrossRef] [PubMed]
Xu J Molday LL Molday RS Sarunic MV . In vivo imaging of the mouse model of X-linked juvenile retinoschisis with fourier domain optical coherence tomography. Invest Ophthalmol Vis Sci. 2009;50:2989–2993. [CrossRef] [PubMed]
Huber G Beck SC Grimm C . Spectral domain optical coherence tomography in mouse models of retinal degeneration. Invest Ophthalmol Vis Sci. 2009;50:5888–5895. [CrossRef] [PubMed]
Nagata A Higashide T Ohkubo S Takeda H Sugiyama K . In vivo quantitative evaluation of the rat retinal nerve fiber layer with optical coherence tomography. Invest Ophthalmol Vis Sci. 2009;50:2809–2815. [CrossRef] [PubMed]
Ruggeri M Wehbe H Jiao S . In vivo three-dimensional high-resolution imaging of rodent retina with spectral-domain optical coherence tomography. Invest Ophthalmol Vis Sci. 2007;48:1808–1814. [CrossRef] [PubMed]
John SW Smith RS Savinova OV . Essential iris atrophy, pigment dispersion, and glaucoma in DBA/2J mice. Invest Ophthalmol Vis Sci. 1998;39:951–962. [PubMed]
Guy J Qi X Koilkonda RD . Efficiency and safety of AAV-mediated gene delivery of the human ND4 complex I subunit in the mouse visual system. Invest Ophthalmol Vis Sci. 2009;50:4205–4214. [CrossRef] [PubMed]
Xiang M Zhou L Macke JP . The Brn-3 family of POU-domain factors: primary structure, binding specificity, and expression in subsets of retinal ganglion cells and somatosensory neurons. J Neurosci. 1995;15:4762–4785. [PubMed]
Mu X Fu X Sun H . Ganglion cells are required for normal progenitor- cell proliferation but not cell-fate determination or patterning in the developing mouse retina. Curr Biol. 2005;15:525–530. [CrossRef] [PubMed]
Schuman JS Hee MR Puliafito CA . Quantification of nerve fiber layer thickness in normal and glaucomatous eyes using optical coherence tomography. Arch Ophthalmol. 1995;113:586–596. [CrossRef] [PubMed]
Schuman JS Pedut-Kloizman T Hertzmark E . Reproducibility of nerve fiber layer thickness measurements using optical coherence tomography. Ophthalmology. 1996;103:1889–1898. [CrossRef] [PubMed]
Wang SW Mu X Bowers WJ . Brn3b/Brn3c double knockout mice reveal an unsuspected role for Brn3c in retinal ganglion cell axon outgrowth. Development. 2002;129:467–477. [PubMed]
Haverkamp S Wassle H . Immunocytochemical analysis of the mouse retina. J Comp Neurol. 2000;424:1–23. [CrossRef] [PubMed]
Lee ES Lee JY Jeon CJ . Types and density of calretinin-containing retinal ganglion cells in mouse. Neurosci Res 2010;66:141–150. [CrossRef] [PubMed]
Nadal-Nicolas FM Jimenez-Lopez M Sobrado-Calvo P . Brn3a as a marker of retinal ganglion cells: qualitative and quantitative time course studies in naive and optic nerve-injured retinas. Invest Ophthalmol Vis Sci. 2009;50:3860–3868. [CrossRef] [PubMed]
Soto I Pease ME Son JL Shi X Quigley HA Marsh-Armstrong N . Retinal ganglion cell loss in a rat ocular hypertension model is sectorial and involves early optic nerve axon loss. Invest Ophthalmol Vis Sci. 2011;52:434–441. [CrossRef] [PubMed]
Anderson MG Hawes NL Trantow CM Chang B John SW . Iris phenotypes and pigment dispersion caused by genes influencing pigmentation. Pigment Cell Melanoma Res. 2008;21:565–578. [CrossRef] [PubMed]
Anderson MG Smith RS Hawes NL . Mutations in genes encoding melanosomal proteins cause pigmentary glaucoma in DBA/2J mice. Nat Genet. 2002;30:81–85. [CrossRef] [PubMed]
Howell GR Libby RT Jakobs TC . Axons of retinal ganglion cells are insulted in the optic nerve early in DBA/2J glaucoma. J Cell Biol. 2007;179:1523–1537. [CrossRef] [PubMed]
Jakobs TC Libby RT Ben Y John SW Masland RH . Retinal ganglion cell degeneration is topological but not cell type specific in DBA/2J mice. J Cell Biol. 2005;171:313–325. [CrossRef] [PubMed]
Panagis L Zhao X Ge Y Ren L Mittag TW Danias J . Gene expression changes in areas of focal loss of retinal ganglion cells in the retina of DBA/2J mice. Invest Ophthalmol Vis Sci 2010;51:2024–2034. [CrossRef] [PubMed]
Reichstein D Ren L Filippopoulos T Mittag T Danias J . Apoptotic retinal ganglion cell death in the DBA/2 mouse model of glaucoma. Exp Eye Res. 2007;84:13–21. [CrossRef] [PubMed]
Schlamp CL Li Y Dietz JA Janssen KT Nickells RW . Progressive ganglion cell loss and optic nerve degeneration in DBA/2J mice is variable and asymmetric. BMC Neurosci. 2006;7:66. [CrossRef] [PubMed]
Danias J Lee KC Zamora MF . 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. [CrossRef] [PubMed]
Masland RH . Neuronal diversity in the retina. Curr Opin Neurobiol. 2001;11:431–436. [CrossRef] [PubMed]
Filippopoulos T Danias J Chen B Podos SM Mittag TW . Topographic and morphologic analyses of retinal ganglion cell loss in old DBA/2NNia mice. Invest Ophthalmol Vis Sci. 2006;47:1968–1974. [CrossRef] [PubMed]
Beck SC Schaeferhoff K Michalakis S . In vivo analysis of cone survival in mice. Invest Ophthalmol Vis Sci 2010;51:493–497. [CrossRef] [PubMed]
Morgan JE Waldock A Jeffery G Cowey A . Retinal nerve fibre layer polarimetry: histological and clinical comparison. Br J Ophthalmol. 1998;82:684–690. [CrossRef] [PubMed]
Feng G Mellor RH Bernstein M . Imaging neuronal subsets in transgenic mice expressing multiple spectral variants of GFP. Neuron. 2000;28:41–51. [CrossRef] [PubMed]
Kim IJ Zhang Y Meister M Sanes JR . Laminar restriction of retinal ganglion cell dendrites and axons: subtype-specific developmental patterns revealed with transgenic markers. J Neurosci 2010;30:1452–1462. [CrossRef] [PubMed]
Yang C . Molecular contrast optical coherence tomography: a review. Photochem Photobiol. 2005;81:215–237. [CrossRef] [PubMed]
Applegate B Yang C Izatt J . Theoretical comparison of the sensitivity of molecular contrast optical coherence tomography techniques. Opt Express. 2005;13:8146–8163. [CrossRef] [PubMed]
Boppart SA Oldenburg AL Xu C Marks DL . Optical probes and techniques for molecular contrast enhancement in coherence imaging. J Biomed Opt. 2005;10:41208. [CrossRef] [PubMed]
Figure 1.
 
Representative comparison of SD-OCT B-scan retinal cross-sections (A, C, E) to hematoxylin and eosin (H&E)-stained retinal histologic sections (B, D, F) from the same mice. All retinal layers can be distinguished in both the SD-OCT images and H&E sections. Retinal layers are labeled as follows: neurofiber layer/retinal ganglion cell layer (NFL/RGCL), inner plexiform layer (IPL), inner nuclear layer (INL), outer plexiform layer (OPL), outer nuclear layer (ONL), photoreceptor inner segment/outer segment (IS/OS), and RPE. The NFL of the Brn3b -/- retina (A and B) is qualitatively thinner than that of the Brn3b +/+ (E and F) and Brn3b +/- (C and D) retinas.
Figure 1.
 
Representative comparison of SD-OCT B-scan retinal cross-sections (A, C, E) to hematoxylin and eosin (H&E)-stained retinal histologic sections (B, D, F) from the same mice. All retinal layers can be distinguished in both the SD-OCT images and H&E sections. Retinal layers are labeled as follows: neurofiber layer/retinal ganglion cell layer (NFL/RGCL), inner plexiform layer (IPL), inner nuclear layer (INL), outer plexiform layer (OPL), outer nuclear layer (ONL), photoreceptor inner segment/outer segment (IS/OS), and RPE. The NFL of the Brn3b -/- retina (A and B) is qualitatively thinner than that of the Brn3b +/+ (E and F) and Brn3b +/- (C and D) retinas.
Figure 2.
 
The representative three-dimensional segmentation of the NFL and IPL of Brn3b -/- (A), Brn3b +/- (B), and Brn3b +/+ (C) retinas yields a retinal thickness map. Thickness loss is uniform across the Brn3b -/- retina. Retinal artery locations in the OCT generated fundus images of the Brn3b -/- (D), Brn3b +/- (E), and Brn3b +/+ (F) mice match well with the retinal arteries seen in the thickness map.
Figure 2.
 
The representative three-dimensional segmentation of the NFL and IPL of Brn3b -/- (A), Brn3b +/- (B), and Brn3b +/+ (C) retinas yields a retinal thickness map. Thickness loss is uniform across the Brn3b -/- retina. Retinal artery locations in the OCT generated fundus images of the Brn3b -/- (D), Brn3b +/- (E), and Brn3b +/+ (F) mice match well with the retinal arteries seen in the thickness map.
Figure 3.
 
Retinal thicknesses and SDs (error bars) were calculated using data from the retinal thickness map. Brn3b -/- mice have thinner retinas and nonoverlapping SDs with phenotypically wild-type (Brn3b +/+ and Brn3b +/-) mice.
Figure 3.
 
Retinal thicknesses and SDs (error bars) were calculated using data from the retinal thickness map. Brn3b -/- mice have thinner retinas and nonoverlapping SDs with phenotypically wild-type (Brn3b +/+ and Brn3b +/-) mice.
Figure 4.
 
(A) Cell counts in the NFL of a Brn3b -/-, Brn3b +/-, and a Brn3b +/+ mouse shows a reduction in cell number in the Brn3b -/- NFL. (B) Averaging the top 75% of the cell counts for all mice shows a reduction in cell numbers in Brn3b -/- versus phenotypically wild-type retinas. Error bars represent the maximum cell count for each retina.
Figure 4.
 
(A) Cell counts in the NFL of a Brn3b -/-, Brn3b +/-, and a Brn3b +/+ mouse shows a reduction in cell number in the Brn3b -/- NFL. (B) Averaging the top 75% of the cell counts for all mice shows a reduction in cell numbers in Brn3b -/- versus phenotypically wild-type retinas. Error bars represent the maximum cell count for each retina.
Figure 5.
 
Representative comparison of flattened Z-stack images of Dapi- (A, B, C), Brn3a- (D, E, F), and calretinin-stained (G, H, I) retinal cross sections from Brn3b -/-, Brn3b +/-, and Brn3b +/+ mice. Retinal layers are labeled as previously indicated. The RGCL of Brn3b -/- retinas (A, D, G) has significantly fewer Brn3a- and calretinin-stained cells than the RGCL of Brn3b +/- (B, E, H) and Brn3b +/+ (C, F, I) retinas.
Figure 5.
 
Representative comparison of flattened Z-stack images of Dapi- (A, B, C), Brn3a- (D, E, F), and calretinin-stained (G, H, I) retinal cross sections from Brn3b -/-, Brn3b +/-, and Brn3b +/+ mice. Retinal layers are labeled as previously indicated. The RGCL of Brn3b -/- retinas (A, D, G) has significantly fewer Brn3a- and calretinin-stained cells than the RGCL of Brn3b +/- (B, E, H) and Brn3b +/+ (C, F, I) retinas.
Figure 6.
 
Scatter plot using NFL cell counts as the independent variable and calculated retinal thickness as the dependent variable. A strong correlation between cell counts and calculated retinal thickness is observed (R 2 = 0.9612).
Figure 6.
 
Scatter plot using NFL cell counts as the independent variable and calculated retinal thickness as the dependent variable. A strong correlation between cell counts and calculated retinal thickness is observed (R 2 = 0.9612).
×
×

This PDF is available to Subscribers Only

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

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

×