Automated quantification of optic nerve axons in primate glaucomatous and normal eyes--method and comparison to semi-automated manual quantification

Retrolaminar Demyelination of Structurally Intact Axons in Nonhuman Primate Experimental Glaucoma

Purpose

To determine if structurally intact, retrolaminar optic nerve (RON) axons are demyelinated in nonhuman primate (NHP) experimental glaucoma (EG).

Methods

Unilateral EG NHPs (n = 3) were perfusion fixed, EG and control eyes were enucleated, and foveal Bruch's membrane opening (FoBMO) 30° sectoral axon counts were estimated. Optic nerve heads were trephined; serial vibratome sections (VSs) were imaged and colocalized to a fundus photograph establishing their FoBMO location. The peripheral neural canal region within n = 5 EG versus control eye VS comparisons was targeted for scanning block-face electron microscopic reconstruction (SBEMR) using micro-computed tomographic reconstructions (µCTRs) of each VS. Posterior laminar beams within each µCTR were segmented, allowing a best-fit posterior laminar surface (PLS) to be colocalized into its respective SBEMR. Within each SBEMR, up to 300 axons were randomly traced until they ended (nonintact) or left the block (intact). For each intact axon, myelin onset was identified and myelin onset distance (MOD) was measured relative to the PLS. For each EG versus control SBEMR comparison, survival analyses compared EG and control MOD.

Results

MOD calculations were successful in three EG and five control eye SBEMRs. Within each SBEMR comparison, EG versus control eye axon loss was -32.9%, -8.3%, and -15.2% (respectively), and MOD was increased in the EG versus control SBEMR (P < 0.0001 for each EG versus control SBEMR comparison). When data from all three EG eye SBEMRs were compared to all five control eye SBEMRs, MOD was increased within the EG eyes.

Conclusions

Structurally intact, RON axons are demyelinated in NHP early to moderate EG. Studies to determine their functional status are indicated.

AxonQuantifier: A semi-automated program for quantifying axonal density from whole-mounted optic nerves

BACKGROUND

Here, we present a semi-automated method for quantifying retinal ganglion cell (RGC) axon density at different distances from the optic nerve crush site using longitudinal, confocal microscopy images taken from whole-mounted optic nerves. This method employs the algorithm AxonQuantifier which operates on the freely available program, ImageJ.

NEW METHOD

To validate this method, seven adult male Long Evans rats underwent optic nerve crush injury followed by in vivo treatment with electric fields of varying strengths for 30 days to produce optic nerves with a wide range of axon densities distal to the optic nerve crush site. Prior to euthanasia, RGC axons were labelled with intravitreal injections of cholera toxin B conjugated to Alexa Fluor 647. After dissection, optic nerves underwent tissue clearing, were whole-mounted, and imaged longitudinally using confocal microscopy.

COMPARISON WITH EXISTING METHODS

Five masked raters quantified RGC axon density at 250, 500, 750, 1000, 1250, 1500, 1750, and 2000 µm distances past the optic nerve crush site for the seven optic nerves manually and using AxonQuantifier. Agreement between these methods was assessed using Bland-Altman plots and linear regression. Inter-rater agreement was assessed using the intra-class coefficient.

RESULTS

Semi-automated quantification of RGC axon density demonstrated improved inter-rater agreement and reduced bias values as compared to manual quantification, while also increasing time efficiency 4-fold. Relative to manual quantification, AxonQuantifier tended to underestimate axon density.

CONCLUSIONS

AxonQuantifier is a reliable and efficient method for quantifying axon density from whole mount optic nerves.

AxoNet 2.0: A Deep Learning-Based Tool for Morphometric Analysis of Retinal Ganglion Cell Axons

Purpose

Assessment of glaucomatous damage in animal models is facilitated by rapid and accurate quantification of retinal ganglion cell (RGC) axonal loss and morphologic change. However, manual assessment is extremely time- and labor-intensive. Here, we developed AxoNet 2.0, an automated deep learning (DL) tool that (i) counts normal-appearing RGC axons and (ii) quantifies their morphometry from light micrographs.

Methods

A DL algorithm was trained to segment the axoplasm and myelin sheath of normal-appearing axons using manually-annotated rat optic nerve (ON) cross-sectional micrographs. Performance was quantified by various metrics (e.g., soft-Dice coefficient between predicted and ground-truth segmentations). We also quantified axon counts, axon density, and axon size distributions between hypertensive and control eyes and compared to literature reports.

Results

AxoNet 2.0 performed very well when compared to manual annotations of rat ON (R2 = 0.92 for automated vs. manual counts, soft-Dice coefficient = 0.81 ± 0.02, mean absolute percentage error in axonal morphometric outcomes < 15%). AxoNet 2.0 also showed promise for generalization, performing well on other animal models (R2 = 0.97 between automated versus manual counts for mice and 0.98 for non-human primates). As expected, the algorithm detected decreased in axon density in hypertensive rat eyes (P ≪ 0.001) with preferential loss of large axons (P < 0.001).

Conclusions

AxoNet 2.0 provides a fast and nonsubjective tool to quantify both RGC axon counts and morphological features, thus assisting with assessing axonal damage in animal models of glaucomatous optic neuropathy.

Translational Relevance

This deep learning approach will increase rigor of basic science studies designed to investigate RGC axon protection and regeneration.

Sigma-1 receptor agonist, (+)-pentazocine, is neuroprotective in a Brown Norway rat microbead model of glaucoma

PURPOSE

Glaucoma is a worldwide leading cause of irreversible blindness. Standard treatments lower intraocular pressure (IOP). Novel treatments to prevent optic nerve (ON) degeneration are needed. Here, we investigate the hypothesis that sigma-1 receptor (S1R) agonist (+)-pentazocine (PTZ) is neuroprotective in a Brown Norway (BN) rat, microbead model of glaucoma.

METHODS

BN rats (9-11 weeks, male and female) were treated by intraperitoneal injection, 3 times per week with (+)-PTZ (2 mg/kg) or vehicle (VEH) alone. Treatment started 1 week prior to intraocular injection of polystyrene microbeads to elevate IOP. IOP was measured 2-3 times per week. Five weeks post microbead injection, rats were euthanized. ONs were removed, then fixed and processed for 63x oil, light microscope imaging of toluidine blue stained ON cross sections. To facilitate comparison of ON morphology from VEH and (+)-PTZ treated rats with similar ocular hypertensive insults, rats were assigned to low (IOP ≤15.8 mmHg), moderate (15.8 < IOP <28.0 mmHg), and high (IOP ≥28.0 mmHg) groups based on average IOP in the microbead injected eye. Axon numbers, axon density, axonal and glial areas, axon loss, and axon size distributions of naïve, bead, and contralateral ONs were assessed using QuPath program for automated image analysis.

RESULTS

(+)-PTZ treatment of BN rats protected ONs from damage caused by moderate IOP elevation. Treatment with (+)-PTZ significantly reduced axon loss and glial areas, and increased axon density and axonal areas compared to ONs from VEH treated rats with moderate IOP. (+)-PTZ-mediated neuroprotection was independent of IOP lowering effects. At average IOP ≥28.0 mmHg, (+)-PTZ treatment did not provide measurable neuroprotection. ONs from contralateral eyes exhibited subtle, complex changes in response to conditions in the bead eyes.

CONCLUSIONS

S1R agonist (+)-PTZ shows promise as a neuroprotective treatment for glaucoma. Future studies to understand the complex molecular mechanisms by which (+)-PTZ provides this neuroprotection are needed.

AxonDeep: Automated Optic Nerve Axon Segmentation in Mice With Deep Learning

Purpose

Optic nerve damage is the principal feature of glaucoma and contributes to vision loss in many diseases. In animal models, nerve health has traditionally been assessed by human experts that grade damage qualitatively or manually quantify axons from sampling limited areas from histologic cross sections of nerve. Both approaches are prone to variability and are time consuming. First-generation automated approaches have begun to emerge, but all have significant shortcomings. Here, we seek improvements through use of deep-learning approaches for segmenting and quantifying axons from cross-sections of mouse optic nerve.

Methods

Two deep-learning approaches were developed and evaluated: (1) a traditional supervised approach using a fully convolutional network trained with only labeled data and (2) a semisupervised approach trained with both labeled and unlabeled data using a generative-adversarial-network framework.

Results

From comparisons with an independent test set of images with manually marked axon centers and boundaries, both deep-learning approaches outperformed an existing baseline automated approach and similarly to two independent experts. Performance of the semisupervised approach was superior and implemented into AxonDeep.

Conclusions

AxonDeep performs automated quantification and segmentation of axons from healthy-appearing nerves and those with mild to moderate degrees of damage, similar to that of experts without the variability and constraints associated with manual performance.

Translational Relevance

Use of deep learning for axon quantification provides rapid, objective, and higher throughput analysis of optic nerve that would otherwise not be possible.

Age and intraocular pressure in murine experimental glaucoma

Age and intraocular pressure (IOP) are the two most important risk factors for the development and progression of open-angle glaucoma. While IOP is commonly considered in models of experimental glaucoma (EG), most studies use juvenile or adult animals and seldom older animals which are representative of the human disease. This paper provides a concise review of how retinal ganglion cell (RGC) loss, the hallmark of glaucoma, can be evaluated in EG with a special emphasis on serial in vivo imaging, a parallel approach used in clinical practice. It appraises the suitability of EG models for the purpose of in vivo imaging and argues for the use of models that provide a sustained elevation of IOP, without compromise of the ocular media. In a study with parallel cohorts of adult (3-month-old, equivalent to 20 human years) and old (2-year-old, equivalent to 70 human years) mice, we compare the effects of elevated IOP on serial ganglion cell complex thickness and individual RGC dendritic morphology changes obtained in vivo. We also evaluate how age modulates the impact of elevated IOP on RGC somal and axonal density in histological analysis as well the density of melanopsin RGCs. We discuss the challenges of using old animals and emphasize the potential of single RGC imaging for understanding the pathobiology of RGC loss and evaluating new therapeutic avenues.

Quantification and image-derived phenotyping of retinal ganglion cell nuclei in the nee mouse model of congenital glaucoma

The nee mouse model exhibits characteristic features of congenital glaucoma, a common cause of childhood blindness. The current study of nee mice had two components. First, the time course of neurodegeneration in nee retinal flat-mounts was studied over time using a retinal ganglion cell (RGC)-marker, BRN3A; a pan-nuclear marker, TO-PRO-3; and H&E staining. Based on segmentation of nuclei using ImageJ and RetFM-J, this analysis identified a rapid loss of BRN3A⁺ nuclei from 4 to 15 weeks of age, with the first statistically significant difference in average density compared to age-matched controls detected in 8-week-old cohorts (49% reduction in nee). Consistent with a model of glaucoma, no reductions in BRN3A^- nuclei were detected, but the combined analysis indicated that some RGCs lost BRN3A marker expression prior to actual cell loss. These results have a practical application in the design of experiments using nee mice to study mechanisms or potential therapies for congenital glaucoma. The second component of the study pertains to a discovery-based analysis of the large amount of image data with 748,782 segmented retinal nuclei. Using the automatedly collected region of interest feature data captured by ImageJ, we tested whether RGC density of glaucomatous mice was significantly correlated to average nuclear area, perimeter, Feret diameter, or MinFeret diameter. These results pointed to two events influencing nuclear size. For variations in RGC density above approximately 3000 nuclei/mm² apparent spreading was observed, in which BRN3A^- nuclei-regardless of genotype-became slightly larger as RGC density decreased. This same spreading occurred in BRN3A⁺ nuclei of wild-type mice. For variation in RGC density below 3000 nuclei/mm², which only occurred in glaucomatous nee mutants, BRN3A⁺ nuclei became smaller as disease was progressively severe. These observations have relevance to defining RGCs of relatively higher sensitivity to glaucomatous cell death and the nuclear dynamics occurring during their demise.

AxoNet: A deep learning-based tool to count retinal ganglion cell axons

In this work, we develop a robust, extensible tool to automatically and accurately count retinal ganglion cell axons in optic nerve (ON) tissue images from various animal models of glaucoma. We adapted deep learning to regress pixelwise axon count density estimates, which were then integrated over the image area to determine axon counts. The tool, termed AxoNet, was trained and evaluated using a dataset containing images of ON regions randomly selected from whole cross sections of both control and damaged rat ONs and manually annotated for axon count and location. This rat-trained network was then applied to a separate dataset of non-human primate (NHP) ON images. AxoNet was compared to two existing automated axon counting tools, AxonMaster and AxonJ, using both datasets. AxoNet outperformed the existing tools on both the rat and NHP ON datasets as judged by mean absolute error, R2 values when regressing automated vs. manual counts, and Bland-Altman analysis. AxoNet does not rely on hand-crafted image features for axon recognition and is robust to variations in the extent of ON tissue damage, image quality, and species of mammal. Therefore, AxoNet is not species-specific and can be extended to quantify additional ON characteristics in glaucoma and potentially other neurodegenerative diseases.

The Effects of Glaucoma on the Pressure-Induced Strain Response of the Human Lamina Cribrosa

Purpose

To measure the ex vivo pressure-induced strain response of the human optic nerve head and analyze for variations with glaucoma diagnosis and optic nerve axon damage.

Methods

The posterior sclera of 16 eyes from 8 diagnosed glaucoma donors and 10 eyes from 6 donors with no history of glaucoma were inflation tested between 5 and 45 mm Hg. The optic nerve from each donor was examined for degree of axon loss. The posterior volume of the lamina cribrosa (LC) was imaged with second harmonic generation and analyzed using volume correlation to calculate LC strains between 5 and 10 and 5 and 45 mm Hg.

Results

Eye length and LC area were larger in eyes diagnosed with glaucoma (P ≤ 0.03). Nasal-temporal EXX and circumferential Eθθ strains were lower in the LC of diagnosed glaucoma eyes at 10 mm Hg (P ≤ 0.05) and 45 mm Hg (P ≤ 0.07). EXX was smaller in the LC of glaucoma eyes with <25% axon loss compared with undamaged normal eyes (P = 0.01, 45 mm Hg). In general, the strains were larger in the peripheral than central LC. The ratio of the maximum principal strain Emax in the peripheral to central LC was larger in glaucoma eyes with >25% axon loss than in glaucoma eyes with milder damage (P = 0.004, 10 mm Hg).

Conclusions

The stiffness of the LC pressure-strain response was greater in diagnosed glaucoma eyes and varied with glaucomatous axon damage. Lower LC strains in glaucoma eyes with milder damage may represent baseline biomechanical behavior that contributes to axon loss, whereas greater LC strain and altered radial LC strain variation in glaucoma eyes with more severe damage may be caused by glaucoma-related remodeling.

QuPath Automated Analysis of Optic Nerve Degeneration in Brown Norway Rats

Purpose

A novel application of QuPath open-source digital analysis software is used to provide in-depth morphological analysis of progressive optic nerve (ON) degeneration in rats.

Methods

QuPath software was adapted to assess axon and gliotic morphology in toluidine blue-stained, Brown Norway rat ON light micrographs. QuPath axon numbers, density, size distributions, and gliotic areas were obtained from test images and ON cross-sections separated by damage grade. QuPath results were compared with manual counting, AxonJ, and electron microscopy axon estimates.

Results

QuPath-derived axon number, density, and diameter decreased with increasing ON damage. Axon density negatively correlated with gliotic areas in test images (R2 = 0.759; P < 0.0001; N = 40) and in ON cross-sections (R2 = 0.803; P < 0.0004; N = 10). Although axon losses occurred across most axon diameters, large axons were more susceptible to degeneration. The exception was swollen axons > 2 µm, which increased in moderately but not severely damaged images. QuPath axon counts correlated strongly with manual counts of test images (R2 = 0.956; P < 0.0001). QuPath outperformed AxonJ on test images and total ON axon counts. Compared to electron microscopy analysis, QuPath undercounted ON axons; however, correlation between the methods was robust (R2 = 0.797; P < 0.001; N = 10).

Conclusions

QuPath analysis reliably identified axon loss, axon morphology changes, and gliotic expansion that occurred in degenerating ONs.

Translational Relevance

QuPath is a valuable tool for rapid, automated, analysis of healthy and degenerating ONs. Reproducible preclinical studies for new glaucoma treatments depend on unbiased in-depth analysis of ON pathology. This was provided by the QuPath approach.

Rapid and Precise Semi-Automatic Axon Quantification in Human Peripheral Nerves

We developed a time-efficient semi-automated axon quantification method using freeware in human cranial nerve sections stained with paraphenylenediamine (PPD). It was used to analyze a total of 1238 facial and masseteric nerve biopsies. The technique was validated by comparing manual and semi-automated quantification of 129 (10.4%) randomly selected biopsies. The software-based method demonstrated a sensitivity of 94% and a specificity of 87%. Semi-automatic axon counting was significantly faster (p < 0.001) than manual counting. It took 1 hour and 47 minutes for all 129 biopsies (averaging 50 sec per biopsy, 0.04 seconds per axon). The counting process is automatic and does not need to be supervised. Manual counting took 21 hours and 6 minutes in total (average 9 minutes and 49 seconds per biopsy, 0.52 seconds per axon). Our method showed a linear correlation to the manual counts (R = 0.944 Spearman rho). Attempts have been made by several research groups to automate axonal load quantification. These methods often require specific hard- and software and are therefore only accessible to a few specialized laboratories. Our semi-automated axon quantification is precise, reliable and time-sparing using publicly available software and should be useful for an effective axon quantification in various human peripheral nerves.

A Novel Tree Shrew (Tupaia belangeri) Model of Glaucoma

Purpose

Primates and rodents are used widely as animal models of glaucoma, but each has significant limitations. Researchers need additional animal models that closely resemble the relevant anatomy and pathologic features of the human disease to more quickly advance research. We validate a novel glaucoma animal model in tree shrews (Tupaia belangeri).

Methods

Experimental glaucoma was induced in adult tree shrews (n = 8) by injecting 50 μL of a 25 mg/mL ferromagnetic bead solution into the anterior chamber. Beads were directed into the iridocorneal angle with a magnet to impede aqueous outflow. Animals were followed for 3 months with weekly IOP measurements and biweekly spectral domain optical coherence tomography (SD-OCT) images of the optic nerve head. Histopathology of the optic nerve and optic nerve axon counts were completed at the end of the study.

Results

The 12-week average mean IOP was 22.7 ± 3.6 and 8.6 ± 2.9 mm Hg in the treated and control eyes, respectively. Longitudinal analysis showed significant retinal nerve fiber layer (RNFL) thinning throughout the study. Axon counts were significantly reduced (59.7%) in treated versus control eyes. SD-OCT imaging showed cupping and posterior displacement of the lamina cribrosa in glaucomatous eyes. RNFL thickness and optic nerve axon counts were reduced consistent with IOP elevation. Optic nerves demonstrated histopathology consistent with glaucomatous optic neuropathy.

Conclusions

Tree shrews with experimental glaucoma show key pathologic characteristics of the human disease. The tree shrew model of glaucoma has the potential to help researchers accelerate our understanding of glaucoma pathophysiology.

An automated method for the quantification of transgene expression in motor axons of the peripheral nerve

BACKGROUND

Determination of transgene expression in motor axons of peripheral nerves is important in evaluating the effectiveness of viral transduction. Currently only manual and semi-automatic methods of quantification have been employed for quantification in immunolabeled nerve sections, but automatic methods exist for axon counting only in brightfield sections. Manual and semi-automatic methods can suffer from inter- and intraobserver bias, sampling bias and can be time consuming to implement.

NEW METHOD

A fully automated method using ImageJ and the Nucleus Counter plugin was developed to quantify the fraction of green fluorescent protein (GFP) labeled acetylcholine transferase positive axons in triple immunolabeled peripheral nerve sections. This method utilizes the Nucleus Counter to generate axonal regions of interest which are quantified for colocalization with GFP expression and nonoverlap with Laminin. Thresholding using histograms generated from control animals is used to remove noise.

RESULTS

The automated method is able to successfully distinguish transgenic GFP expressing mice from wild type. Using computer generated peripheral nerve sections, the automated method has less than 5% error at signal-to-noise ratios greater than 10% of baseline.

COMPARISONS WITH EXISTING METHODS

This method has comparable performance in false positive rates (<1%) and a 95% predictive interval that closely matches existing fully automated methods for quantification in brightfield sections. It outperforms the intra- and interobserver differences of manual and semi-automated methods for quantification.

CONCLUSIONS

This automated quantification method provides a fast and robust means of determining the fraction of labeled axons in peripheral nerve sections.

3D Histomorphometric Reconstruction and Quantification of the Optic Nerve Head Connective Tissues

Accurately characterizing the 3D geometry of the optic nerve head neural and connective tissues has been the goal of a large and important body of scientific work. In the present report, we summarize our methods for the high-resolution, digital, 3D histomorphometric reconstruction of the optic nerve head tissues, including their visualization, parameterization, and quantification. In addition, we present our methods for between-eye comparisons of this anatomy, and their use to determine animal-specific and experiment-wide experimental glaucoma versus Control eye differences in the unilateral, monkey experimental glaucoma model. Finally, we demonstrate its application to finite element modeling, 3D optic nerve head reconstruction of other species, and 3D optic nerve head reconstructions using other imaging modalities.

Axon and Myelin Morphology in Animal and Human Spinal Cord

Characterizing precisely the microstructure of axons, their density, size and myelination is of interest for the neuroscientific community, for example to help maximize the outcome of studies on white matter (WM) pathologies of the spinal cord (SC). The existence of a comprehensive and structured database of axonal measurements in healthy and disease models could help the validation of results obtained by different researchers. The purpose of this article is to provide such a database of healthy SC WM, to discuss the potential sources of variability and to suggest avenues for robust and accurate quantification of axon morphometry based on novel acquisition and processing techniques. The article is organized in three sections. The first section reviews morphometric results across species according to range of densities and counts of myelinated axons, axon diameter and myelin thickness, and characteristics of unmyelinated axons in different regions. The second section discusses the sources of variability across studies, such as age, sex, spinal pathways, spinal levels, statistical power and terminology in regard to tracts and protocols. The third section presents new techniques and perspectives that could benefit histology studies. For example, coherent anti-stokes Raman spectroscopy (CARS) imaging can provide sub-micrometric resolution without the need for fixation and staining, while slide scanners and stitching algorithms can provide full cross-sectional area of SC. In combination with these acquisition techniques, automatic segmentation algorithms for delineating axons and myelin sheath can help provide large-scale statistics on axon morphometry.

Racioethnic Differences in Human Posterior Scleral and Optic Nerve Stump Deformation

Purpose

The purpose of this study was to quantify the biomechanical response of human posterior ocular tissues from donors of various racioethnic groups to better understand how differences in these properties may play a role in the racioethnic health disparities known to exist in glaucoma.

Methods

Sequential digital image correlation (S-DIC) was used to measure the pressure-induced surface deformations of 23 normal human posterior poles from three racioethnic groups: African descent (AD), European descent (ED), and Hispanic ethnicity (HIS). Regional in-plane principal strains were compared across three zones: the optic nerve stump (ONS), the peripapillary (PP) sclera, and non-PP sclera.

Results

The PP scleral tensile strains were found to be lower for ED eyes compared with AD and HIS eyes at 15 mm Hg (P = 0.024 and 0.039, respectively). The mean compressive strains were significantly higher for AD eyes compared with ED eyes at 15 mm Hg (P = 0.018). We also found that the relationship between tensile strain and pressure was significant for those of ED and HIS eyes (P < 0.001 and P = 0.004, respectively), whereas it was not significant for those of AD (P = 0.392).

Conclusions

Our results suggest that, assuming glaucomatous nerve loss is caused by mechanical strains in the vicinity of the optic nerve head, the mechanism of increased glaucoma prevalence may be different in those of AD versus HIS. Our ONS strain analysis also suggested that it may be important to account for ONS geometry and material properties in future scleral biomechanical analysis.

The connective tissue phenotype of glaucomatous cupping in the monkey eye - Clinical and research implications

In a series of previous publications we have proposed a framework for conceptualizing the optic nerve head (ONH) as a biomechanical structure. That framework proposes important roles for intraocular pressure (IOP), IOP-related stress and strain, cerebrospinal fluid pressure (CSFp), systemic and ocular determinants of blood flow, inflammation, auto-immunity, genetics, and other non-IOP related risk factors in the physiology of ONH aging and the pathophysiology of glaucomatous damage to the ONH. The present report summarizes 20 years of technique development and study results pertinent to the characterization of ONH connective tissue deformation and remodeling in the unilateral monkey experimental glaucoma (EG) model. In it we propose that the defining pathophysiology of a glaucomatous optic neuropathy involves deformation, remodeling, and mechanical failure of the ONH connective tissues. We view this as an active process, driven by astrocyte, microglial, fibroblast and oligodendrocyte mechanobiology. These cells, and the connective tissue phenomena they propagate, have primary and secondary effects on retinal ganglion cell (RGC) axon, laminar beam and retrolaminar capillary homeostasis that may initially be "protective" but eventually lead to RGC axonal injury, repair and/or cell death. The primary goal of this report is to summarize our 3D histomorphometric and optical coherence tomography (OCT)-based evidence for the early onset and progression of ONH connective tissue deformation and remodeling in monkey EG. A second goal is to explain the importance of including ONH connective tissue processes in characterizing the phenotype of a glaucomatous optic neuropathy in all species. A third goal is to summarize our current efforts to move from ONH morphology to the cell biology of connective tissue remodeling and axonal insult early in the disease. A final goal is to facilitate the translation of our findings and ideas into neuroprotective interventions that target these ONH phenomena for therapeutic effect.

Comparing Optic Nerve Head Rim Width, Rim Area, and Peripapillary Retinal Nerve Fiber Layer Thickness to Axon Count in Experimental Glaucoma

Purpose

We compare spectral-domain optical coherence tomography (SDOCT) measurements of minimum rim width (MRW), minimum rim area (MRA), and peripapillary retinal nerve fiber layer thickness (RNFLT) to complete orbital optic nerve axon counts in nonhuman primates (NHP) with unilateral experimental glaucoma (EG).

Methods

Biweekly SDOCT measurements of MRW, MRA, and RNFLT were acquired under manometric IOP control (10 mm Hg) in 51 NHP during baseline (mean ± SD, 5.0 ± 1.6 sessions) and after laser photocoagulation was applied to the trabecular meshwork of one eye to induce chronic IOP elevation. At the study endpoint (predefined for each NHP), 100% axon counts were obtained from each optic nerve.

Results

For SDOCT parameters at baseline, the correlation between the two eyes of each animal was strongest for RNFLT (R = 0.97) and MRW (R = 0.97), but lower for MRA (R = 0.85). At the final time point, average values in EG eyes relative to control eyes were: −22% for RNFLT, −38% for MRW, −36% for MRA, and −36% for optic nerve axons. The correlation with axon counts was strongest for RNFLT (R = 0.81), compared to MRW (R = 0.72, P = 0.001) or MRA (R = 0.70, P = 0.001). Diagnostic sensitivity was 75% for RNFLT, 90% for MRW, and 88% for MRA; all had 100% specificity.

Conclusions

Peripapillary RNFLT was correlated more closely with total orbital optic nerve axon count than were the ONH parameters MRW or MRA. This is likely because glaucomatous deformation (beyond axon loss alone) has a greater influence on the ONH parameters MRW and MRA than on RNFLT.

Cupping in the Monkey Optic Nerve Transection Model Consists of Prelaminar Tissue Thinning in the Absence of Posterior Laminar Deformation

Purpose

To use optical coherence tomography (OCT) to test the hypothesis that optic nerve head (ONH) “cupping” in the monkey optic nerve transection (ONT) model does not include posterior laminar deformation.

Methods

Five monkeys (aged 5.5–7.8 years) underwent ONH and retinal nerve fiber layer (RNFL) OCT imaging five times at baseline and biweekly following unilateral ONT until euthanization at ∼40% RNFL loss. Retinal nerve fiber layer thickness (RNFLT) and minimum rim width (MRW) were calculated from each pre- and post-ONT imaging session. The anterior lamina cribrosa surface (ALCS) was delineated within baseline and pre-euthanasia data sets. Significant ONT versus control eye pre-euthanasia change in prelaminar tissue thickness (PLTT), MRW, RNFLT, and ALCS depth (ALCSD) was determined using a linear mixed-effects model. Eye-specific change in each parameter exceeded the 95% confidence interval constructed from baseline measurements.

Results

Animals were euthanized 49 to 51 days post ONT. Overall ONT eye change from baseline was significant for MRW (−26.2%, P = 0.0011), RNFLT (−43.8%, P < 0.0001), PLTT (−23.8%, P = 0.0013), and ALCSD (−20.8%, P = 0.033). All five ONT eyes demonstrated significant eye-specific decreases in MRW (−23.7% to −31.8%) and RNFLT (−39.6% to −49.7%). Four ONT eyes showed significant PLTT thinning (−23.0% to −28.2%). The ALCS was anteriorly displaced in three of the ONT eyes (−25.7% to −39.2%). No ONT eye demonstrated posterior laminar displacement.

Conclusions

Seven weeks following surgical ONT in the monkey eye, ONH cupping involves prelaminar and rim tissue thinning without posterior deformation of the lamina cribrosa.

Lamina Cribrosa Microarchitecture in Monkey Early Experimental Glaucoma: Global Change

Purpose

The purpose of this study was to characterize experimental glaucoma (EG) versus control eye differences in lamina cribrosa (LC), beam diameter (BD), pore diameter (PD), connective tissue volume fraction (CTVF), connective tissue volume (CTV), and LC volume (LV) in monkey early EG.

Methods

Optic nerve heads (ONHs) of 14 unilateral EG and 6 bilateral normal (BN) monkeys underwent three-dimensional reconstruction and LC beam segmentation. Each beam and pore voxel was assigned a diameter based on the largest sphere that contained it before transformation to a common cylinder with inner, middle, and outer layers. Full-thickness and layer averages for BD, PD, CTVF, CTV, and LV were calculated for each ONH. Beam diameter and PD distributions for each ONH were fit to a gamma distribution and summarized by scale and shape parameters. Experimental glaucoma and depth effects were assessed for each parameter by linear mixed-effects (LME) modeling. Animal-specific EG versus control eye differences that exceeded the maximum intereye difference among the six BN animals were considered significant.

Results

Overall EG eye mean PD was 12.8% larger (28.2 ± 5.6 vs. 25.0 ± 3.3 μm), CTV was 26.5% larger (100.06 ± 47.98 vs. 79.12 ± 28.35 × 10⁶ μm³), and LV was 40% larger (229.29 ± 98.19 vs. 163.63 ± 39.87 × 10⁶ μm³) than control eyes (P ≤ 0.05, LME). Experimental glaucoma effects were significantly different by layer for PD (P = 0.0097) and CTVF (P < 0.0001). Pore diameter expanded consistently across all PDs. Experimental glaucoma eye-specific parameter change was variable in magnitude and direction.

Conclusions

Pore diameter, CTV, and LV increase in monkey early EG; however, EG eye-specific change is variable and includes both increases and decreases in BD and CTVF.

Automated Axon Counting in Rodent Optic Nerve Sections with AxonJ

We have developed a publicly available tool, AxonJ, which quantifies the axons in optic nerve sections of rodents stained with paraphenylenediamine (PPD). In this study, we compare AxonJ's performance to human experts on 100x and 40x images of optic nerve sections obtained from multiple strains of mice, including mice with defects relevant to glaucoma. AxonJ produced reliable axon counts with high sensitivity of 0.959 and high precision of 0.907, high repeatability of 0.95 when compared to a gold-standard of manual assessments and high correlation of 0.882 to the glaucoma damage staging of a previously published dataset. AxonJ allows analyses that are quantitative, consistent, fully-automated, parameter-free, and rapid on whole optic nerve sections at 40x. As a freely available ImageJ plugin that requires no highly specialized equipment to utilize, AxonJ represents a powerful new community resource augmenting studies of the optic nerve using mice.

The Connective Tissue Components of Optic Nerve Head Cupping in Monkey Experimental Glaucoma Part 1: Global Change

PURPOSE

To characterize optic nerve head (ONH) connective tissue change within 21 monkey experimental glaucoma (EG) eyes, so as to identify its principal components.

METHODS

Animals were imaged three to five times at baseline then every 2 weeks following chronic unilateral IOP elevation, and euthanized early through end-stage confocal scanning laser tomographic change. Optic nerve heads were serial-sectioned, three-dimensionally (3D) reconstructed, delineated, and quantified. Overall EG versus control eye differences were assessed by general estimating equations (GEE). Significant, animal-specific, EG eye change was required to exceed the maximum physiologic intereye differences in six healthy animals.

RESULTS

Overall EG eye change was significant (P < 0.0026) and animal-specific EG eye change most frequent, for five phenomena (number of EG eyes and range of animal-specific change): posterior laminar deformation (21, -29 to -437 μm), laminar thickening (11, 20-73 μm) and thinning (3, -23 to -31 μm), scleral canal expansion (17, 20-139 μm), outward anterior (16, -16 to -124 μm) and posterior (17, -22 to -279 μm) laminar insertion migration, and peripapillary scleral bowing (11, 21-77 μm). Experimental glaucoma versus control eye laminar thickness differences were bimodal in behavior, being thickened in most EG eyes demonstrating the least deformation and less thickened or thinned in most EG eyes demonstrating the greatest deformation.

CONCLUSIONS

Our postmortem studies retrospectively identify five connective tissue components of ONH "cupping" in monkey EG which serve as targets for longitudinally staging and phenotyping ONH connective tissue alteration within all forms of monkey and human optic neuropathy.

Compromised Optic Nerve Blood Flow and Autoregulation Secondary to Neural Degeneration

PURPOSE

To test the hypothesis that optic nerve head (ONH) blood flow (BF) and autoregulation compromise are consequences of optic nerve degeneration induced by surgical optic nerve transection (ONT).

METHODS

In both eyes of five nonhuman primates, peripapillary retinal nerve fiber layer thickness (RNFLT) was measured by spectral-domain optical coherence tomography. Optic nerve head BF and dynamic autoregulation responses to a rapid manometric IOP increase (from 10-40 mm Hg) were measured by Laser Speckle Flowgraphy. The measurements were conducted every 10 to 15 days before and after unilateral ONT. Post-ONT measurements were repeated until RNFLT in the ONT eye was reduced by more than 40% of baseline value.

RESULTS

After ONT, RNFLT, and ONH BF progressively declined over time (P < 0.0001 and P = 0.02, respectively). Longitudinal changes between the two were highly correlated (P < 0.0001). When data was grouped by test session, the first significant decreases for RNFLT and BF were found 13 ± 0.8 and 24 ± 3.2 days post ONT, respectively (P < 0.05, both). At the final time point (55 ± 0.5 days post ONT), RNFLT, and BF were reduced by 44% ± 2.0% and 38 ± 5.0% from baseline, respectively. Dynamic autoregulation analysis showed marginal increased response time in post-ONT eyes (P = 0.05). Control eyes showed no longitudinal changes for any parameter.

CONCLUSIONS

The close association between RNFLT loss and ONH BF decrease following optic nerve degeneration demonstrated a clear cause and effect relationship. Increased BF response time appears to be a sign of dynamic autoregulation dysfunction in this ONT model.

Glial coverage in the optic nerve expands in proportion to optic axon loss in chronic mouse glaucoma

Within the white matter, axonal loss by neurodegeneration is coupled to glial cell changes in gene expression, structure and function commonly termed gliosis. Recently, we described the highly variable expansion of gliosis alebosco@neuro.utah.edu in degenerative optic nerves from the DBA/2J mouse model of chronic, age-related glaucoma. Here, to estimate and compare the levels of axonal loss with the expansion of glial coverage and axonal degeneration in DBA/2J nerves, we combined semiautomatic axon counts with threshold-based segmentation of total glial/scar areas and degenerative axonal profiles in plastic cross-sections. In nerves ranging from mild to severe degeneration, we found that the progression of axonal dropout is coupled to an increase of gliotic area. We detected a strong correlation between axon loss and the aggregate coverage by glial cells and scar, whereas axon loss did not correlate with the small fraction of degenerating profiles. Nerves with low to medium levels of axon loss displayed moderate glial reactivity, consisting of hypertrophic astrocytes, activated microglia and normal distribution of oligodendrocytes, with minimal reorganization of the tissue architecture. In contrast, nerves with extensive axonal loss showed prevalent rearrangement of the nerve, with loss of axon fascicle territories and enlarged or almost continuous gliotic and scar domains, containing reactive astrocytes, oligodendrocytes and activated microglia. These findings support the value of optic nerve gliotic expansion as a quantitative estimate of optic neuropathy that correlates with axon loss, applicable to grade the severity of optic nerve damage in mouse chronic glaucoma.

Relating Retinal Ganglion Cell Function and Retinal Nerve Fiber Layer (RNFL) Retardance to Progressive Loss of RNFL Thickness and Optic Nerve Axons in Experimental Glaucoma

PURPOSE

To relate changes in retinal function and retinal nerve fiber layer (RNFL) retardance to loss of RNFL thickness and optic nerve axon counts in a nonhuman primate (NHP) model of experimental glaucoma (EG).

METHODS

Bilateral longitudinal measurements of peripapillary RNFL thickness (spectral-domain optical coherence tomography, SDOCT; Spectralis), retardance (GDxVCC), and multifocal electroretinography (mfERG; VERIS) were performed in 39 NHP at baseline (BL; median, 5 recordings; range, 3-10) and weekly after induction of unilateral EG by laser photocoagulation of the trabecular meshwork. Multifocal ERG responses were high-pass filtered (>75 Hz) to measure high- and low-frequency component (HFC and LFC) amplitudes, including LFC features N1, P1, and N2. High-frequency component amplitudes are known to specifically reflect retinal ganglion cell (RGC) function. Complete (100%) axon counts of orbital optic nerves were obtained in 31/39 NHP.

RESULTS

Postlaser follow-up was 10.4 ± 7.9 months; mean and peak IOP were 18 ± 5 and 41 ± 11 mm Hg in EG eyes, 11 ± 2 and 18 ± 6 mm Hg in control (CTL) eyes. At the final available time point, RNFL thickness had decreased from BL by 14 ± 14%, retardance by 20 ± 11%, and the mfERG HFC by 30 ± 17% (P < 0.0001 each). Longitudinal changes in retardance and HFC were linearly related to RNFL thickness change (R2 = 0.51, P < 0.0001 and R2 = 0.22, P = 0.002, respectively); LFC N2 was weakly related but N1 or P2 (N1: R2 = 0.07, P = 0.11; P1: R2 = 0.04, P = 0.24; N2: R2 = 0.13, P = 0.02). At zero change from BL for RNFL thickness (Y-intercept), retardance was reduced by 11% (95% confidence interval [CI]: -15.3% to -6.8%) and HFC by 21.5% (95% CI: -28.7% to -14.3%). Relative loss of RNFL thickness, retardance, and HFC (EG:CTL) were each related to axon loss (R2 = 0.66, P < 0.0001; R2 = 0.42, P < 0.0001; R2 = 0.42, P < 0.0001, respectively), but only retardance and HFC were significantly reduced at zero relative axon loss (Y-intercept; retardance: -9.4%, 95% CI: -15.5% to -3.4%; HFC: -10.9%, 95% CI: -18.6% to -3.2%; RNFL thickness: +1.8%, 95% CI: -4.9% to +5.4%).

CONCLUSIONS

Retinal nerve fiber layer retardance and RGC function exhibit progressive loss from baseline before any loss of RNFL thickness or orbital optic nerve axons occurs in NHP EG. These in vivo measures might serve as potential biomarkers of early-stage glaucomatous damage preceding axon loss and RGC death.

Evaluation of a Method for Estimating Retinal Ganglion Cell Counts Using Visual Fields and Optical Coherence Tomography

PURPOSE

To evaluate the accuracy and generalizability of a published model that derives estimates of retinal ganglion cell (RGC) counts and relates structural and functional changes due to glaucoma.

METHODS

Both the Harwerth et al. nonlinear model (H-NLM) and the Hood and Kardon linear model (HK-LM) were applied to an independent dataset of frequency-domain optical coherence tomography and visual fields, consisting of 48 eyes of 48 healthy controls, 100 eyes of 77 glaucoma patients and suspects, and 18 eyes of 14 nonarteritic anterior ischemic optic neuropathy (ION) patients with severe vision loss. Using the coefficient of determination R2, the models were compared while keeping constant the topographic maps, specifically a map by Garway-Heath et al. and a separate map by Harwerth et al., which relate sensitivity test stimulus locations to corresponding regions around the optic disc. Additionally, simulations were used to evaluate the assumptions of the H-NLM.

RESULTS

Although the predictions of the HK-LM with the anatomically-derived Garway-Heath et al. map were reasonably good (R2 = 0.31-0.64), the predictions of the H-NLM were poor (R2 < 0) regardless of the map used. Furthermore, simulations of the H-NLM yielded results that differed substantially from RGC estimates based on histology from human subjects. Finally, the value-added of factors increasing the relative complexity of the H-NLM, such as assumptions regarding age- and stage-dependent corrections to structural measures, was unclear.

CONCLUSIONS

Several of the assumptions underlying the H-NLM should be revisited. Studies and models relying on the RGC estimates of the H-NLM should be interpreted with caution.

The non-human primate experimental glaucoma model

The purpose of this report is to summarize the current strengths and weaknesses of the non-human primate (NHP) experimental glaucoma (EG) model through sections devoted to its history, methods, important findings, alternative optic neuropathy models and future directions. NHP EG has become well established for studying human glaucoma in part because the NHP optic nerve head (ONH) shares a close anatomic association with the human ONH and because it provides the only means of systematically studying the very earliest visual system responses to chronic intraocular pressure (IOP) elevation, i.e. the conversion from ocular hypertension to glaucomatous damage. However, NHPs are impractical for studies that require large animal numbers, demonstrate spontaneous glaucoma only rarely, do not currently provide a model of the neuropathy at normal levels of IOP, and cannot easily be genetically manipulated, except through tissue-specific, viral vectors. The goal of this summary is to direct NHP EG and non-NHP EG investigators to the previous, current and future accomplishment of clinically relevant knowledge in this model.

Assessment of retinal ganglion cell damage in glaucomatous optic neuropathy: Axon transport, injury and soma loss

Glaucoma is a disease characterized by progressive axonal pathology and death of retinal ganglion cells (RGCs), which causes structural changes in the optic nerve head and irreversible vision loss. Several experimental models of glaucomatous optic neuropathy (GON) have been developed, primarily in non-human primates and, more recently and commonly, in rodents. These models provide important research tools to study the mechanisms underlying glaucomatous damage. Moreover, experimental GON provides the ability to quantify and monitor risk factors leading to RGC loss such as the level of intraocular pressure, axonal health and the RGC population. Using these experimental models we are able to gain a better understanding of GON, which allows for the development of potential neuroprotective strategies. Here we review the advantages and disadvantages of the relevant and most often utilized methods for evaluating axonal degeneration and RGC loss in GON. Axonal pathology in GON includes functional disruption of axonal transport (AT) and structural degeneration. Horseradish peroxidase (HRP), rhodamine-B-isothiocyanate (RITC) and cholera toxin-B (CTB) fluorescent conjugates have proven to be effective reporters of AT. Also, immunohistochemistry (IHC) for endogenous AT-associated proteins is often used as an indicator of AT function. Similarly, structural degeneration of axons in GON can be investigated via changes in the activity and expression of key axonal enzymes and structural proteins. Assessment of axonal degeneration can be measured by direct quantification of axons, qualitative grading, or a combination of both methods. RGC loss is the most frequently quantified variable in studies of experimental GON. Retrograde tracers can be used to quantify RGC populations in rodents via application to the superior colliculus (SC). In addition, in situ IHC for RGC-specific proteins is a common method of RGC quantification used in many studies. Recently, transgenic mouse models that express fluorescent proteins under the Thy-1 promoter have been examined for their potential to provide specific and selective labeling of RGCs for the study of GON. While these methods represent important advances in assessing the structural and functional integrity of RGCs, each has its advantages and disadvantages; together they provide an extensive toolbox for the study of GON.

Alterations in molecular pathways in the retina of early experimental glaucoma eyes

Glaucoma is a multifactorial, neurodegenerative disease. The molecular mechanisms that underlie the pathophysiological changes in glaucomatous eyes, especially at the early stage of the disease, are poorly understood. Here, we report the findings from a quantitative proteomic analysis of retinas from experimental glaucoma (EG) eyes. An early stage of EG was modeled on unilateral eyes of five nonhuman primates (NHP) by laser treatment-induced elevation of intraocular pressure (IOP). Retinal proteins were extracted from individual EG eyes and their contralateral control eyes of the same animals, respectively, and analyzed by quantitative mass spectrometry (MS). As a result, a total, 475 retinal proteins were confidently identified and quantified. Results of bioinformatic analysis of proteins that showed an increase in the EG eyes suggested changes in apoptosis, DNA damage, immune response, cytoskeleton rearrangement and cell adhesion processes. Interestingly, hemoglobin subunit alpha (HBA) and Ras related C3 botulinum toxin substrate 1 (Rac1) were among the increased proteins. Results of molecular modeling of HBA- and Rac1-associated signaling network implicated the involvement of Mitogen-Activated Protein Kinase (MAPK) pathway in the EG, through which Rac1 may exert a regulatory role on HBA. This is the first observation of this potentially novel signaling network in the NHP retina and in EG. Results of Western blot analyses for Rac1, HBA and a selected MAPK pathway protein indicated synergistic changes in all three proteins in the EG eyes. Further, results of hierarchical cluster analysis of proteomes of control eyes revealed a clear age-proteome relationship, and such relationship appeared disrupted in the EG eyes. In conclusion, our results suggested an increased presence of a potentially novel signaling network at the early stage of glaucoma, and age might be one of the determinant factors in retinal proteomic characteristics under normal conditions.

Semi-automated counting method of axons in transmission electron microscopic images

OBJECTIVE

To evaluate the accuracy of a new semi-automated method for counting axons in transmission electron microscopic (TEM) images.

PROCEDURES

Optic nerve cross sections were obtained from both eyes of Sprague Dawley rats after unilateral induction of chronic ocular hypertension. TEM images (3000× magnification) of cross sections were evaluated by both semi-automated and manual counting methods. The semi-automated counting method was performed using ImageJ software after simple image optimization, and the resulting estimate of axon damage was compared with semiquantitative damage grading scale from light microscopic (LM) images.

RESULTS

Axon counts obtained from the semi-automated method were strongly correlated with those obtained from the manual counting method (Pearson's correlation coefficient r = 0.996, P < 0.001) and from the full manual count from LM images (Spearman's ρ = 0.973, P < 0.001). The semi-automated method measured axonal damage with an error of 0.94 ± 3.16% (mean ± standard deviation), with worse axonal damage associated with greater error. Interobserver and intra-observer variability in axons counts were low (Spearman's ρ = 0.999, P < 0.005). The results of the semi-automated counting method were highly correlated with semiquantitative damage grading scale (Spearman's ρ = 0.965, P < 0.001).

CONCLUSION

Results of our semi-automated method for counting axons in TEM images were strongly correlated with those of conventional counting methods and showed excellent reproducibility.

Comparison of retinal nerve fiber layer thickness in vivo and axonal transport after chronic intraocular pressure elevation in young versus older rats

PURPOSE

To compare in young and old rats longitudinal measurements of retinal nerve fiber layer thickness (RNFLT) and axonal transport 3-weeks after chronic IOP elevation.

METHOD

IOP was elevated unilaterally in 2- and 9.5-month-old Brown-Norway rats by intracameral injections of magnetic microbeads. RNFLT was measured by spectral domain optical coherence tomography. Anterograde axonal transport was assessed from confocal scanning laser ophthalmolscopy of superior colliculi (SC) after bilateral intravitreal injections of cholera toxin-B-488. Optic nerve sections were graded for damage.

RESULTS

Mean IOP was elevated in both groups (young 37, old 38 mmHg, p = 0.95). RNFL in young rats exhibited 10% thickening at 1-week (50.9±8.1 µm, p<0.05) vs. baseline (46.4±2.4 µm), then 7% thinning at 2-weeks (43.0±7.2 µm, p>0.05) and 3-weeks (43.5±4.4 µm, p>0.05), representing 20% loss of dynamic range. RNFLT in old rats showed no significant change at 1-week (44.9±4.1 µm) vs. baseline (49.2±5.3 µm), but progression to 22% thinning at 2-weeks (38.0±3.7 µm, p<0.01) and 3-weeks (40.0±6.6 µm, p<0.05), representing 59% loss of dynamic range. Relative SC fluorescence intensity was reduced in both groups (p<0.001), representing 77-80% loss of dynamic range and a severe transport deficit. Optic nerves showed 75-95% damage (p<0.001). There was greater RNFL thinning in old rats (p<0.05), despite equivalent IOP insult, transport deficit and nerve damage between age groups (all p>0.05).

CONCLUSION

Chronic IOP elevation resulted in severely disrupted axonal transport and optic nerve axon damage in all rats, associated with mild RNFL loss in young rats but a moderate RNFL loss in old rats despite the similar IOP insult. Hence, the glaucomatous injury response within the RNFL depends on age.

Age-related differences in longitudinal structural change by spectral-domain optical coherence tomography in early experimental glaucoma

PURPOSE

To characterize age-related differences in the magnitude of spectral-domain optical coherence tomography (SD-OCT) structural change in early experimental glaucoma (EG).

METHODS

Both eyes from four young (1.4-2.6 years) and four old (18.6-21.9 years) rhesus monkeys were imaged at least three times at baseline, and then every 2 weeks after laser-induced, chronic, unilateral IOP elevation until the onset of EG (confocal scanning laser tomographic surface change confirmed twice). Two to 20 weeks after EG onset, animals were euthanized and optic nerve axon counts for all eyes were performed. Masked operators delineated retinal and ONH landmarks in 40 radial B-scans from each eye and imaging session to quantify change from baseline in five SD-OCT neural and connective tissue parameters. The effects of EG, age, and EG × age interactions on the magnitude, rate (magnitude per postlaser time), and IOP responsiveness (magnitude per cumulative IOP insult) of postlaser parameter change were individually assessed using general estimating equation models.

RESULTS

Presac SD-OCT RNFLT and minimum rim width change and postmortem axon loss was not significantly different in old compared with young EG eyes. The rate of change and IOP responsiveness of the parameters anterior lamina cribrosa surface depth relative to Bruch's membrane opening (BMO) and BMO depth relative to peripheral Bruch's membrane were significantly lower (P < 0.05) in the old compared with the young EG eyes.

CONCLUSIONS

At similar postlaser times, levels of cumulative IOP insult and axonal damage, SD-OCT-detected ONH connective tissue structural change is greater in young compared with old monkey EG eyes.

The Effect of Age on Optic Nerve Axon Counts, SDOCT Scan Quality, and Peripapillary Retinal Nerve Fiber Layer Thickness Measurements in Rhesus Monkeys

PURPOSE

To evaluate the effect of age on optic nerve axon counts, spectral-domain optical coherence tomography (SDOCT) scan quality, and peripapillary retinal nerve fiber layer thickness (RNFLT) measurements in healthy monkey eyes.

METHODS

In total, 83 healthy rhesus monkeys were included in this study (age range: 1.2-26.7 years). Peripapillary RNFLT was measured by SDOCT. An automated algorithm was used to count 100% of the axons and measure their cross-sectional area in postmortem optic nerve tissue samples (N = 46). Simulation experiments were done to determine the effects of optical changes on measurements of RNFLT. An objective, fully-automated method was used to measure the diameter of the major blood vessel profiles within each SDOCT B-scan.

RESULTS

Peripapillary RNFLT was negatively correlated with age in cross-sectional analysis (P < 0.01). The best-fitting linear model was RNFLT_(μm) = -0.40 × age_(years) + 104.5 μm (R² = 0.1, P < 0.01). Age had very little influence on optic nerve axon count; the result of the best-fit linear model was axon count = -1364 × Age_(years) + 1,210,284 (R² < 0.01, P = 0.74). Older eyes lost the smallest diameter axons and/or axons had an increased diameter in the optic nerve of older animals. There was an inverse correlation between age and SDOCT scan quality (R = -0.65, P < 0.0001). Simulation experiments revealed that approximately 17% of the apparent cross-sectional rate of RNFLT loss is due to reduced scan quality associated with optical changes of the aging eye. Another 12% was due to thinning of the major blood vessels.

CONCLUSIONS

RNFLT declines by 4 μm per decade in healthy rhesus monkey eyes. This rate is approximately three times faster than loss of optic nerve axons. Approximately one-half of this difference is explained by optical degradation of the aging eye reducing SDOCT scan quality and thinning of the major blood vessels.

TRANSLATIONAL RELEVANCE

Current models used to predict retinal ganglion cell losses should be reconsidered.

Quantifying optic nerve axons in a cat glaucoma model by a semi-automated targeted counting method

PURPOSE

To describe and validate a semi-automated targeted sampling (SATS) method for quantifying optic nerve axons in a feline glaucoma model.

METHODS

Optic nerve cross sections were obtained from 15 cats, nine with mild to severe glaucoma and six with normal eyes. Optic nerves were dissected, fixed in paraformaldehyde and glutaraldehyde, and processed for light microscopy by resin embedding, sectioning, and staining of axon myelin sheaths with 1% p-phenylenediamine before axon quantification. Commercially available image analysis software was used as a semi-automated axon counting tool (SCT) and was first validated by comparison with a manual axon count (MAC). This counting tool was then used in a SATS method performed by three masked raters and in a semi-automated full count (SAFC) method performed by a single observer. Correlation was assessed between the SCT and MAC using a linear model and analysis of covariance (ANCOVA). Correlation between the SATS and SAFC methods was calculated and the bias, systematic errors, and variance component assessed. The intraclass correlation coefficient (ICC) was determined to establish inter-rater agreement. In addition, the time required to perform the SATS and SAFC methods was evaluated.

RESULTS

Correlation between the axon counts obtained by the SCT and MAC was strong (r = 0.9985). There was evidence of an overcounting of axons by the SCT compared to the MAC with a percentage error rate of 13.0% (95% confidence interval [CI] 11.0%, 15.1%). Both the correlation of SATS count (average per rater) to SAFC (r = 0.9891) and inter-rater agreement (ICC = 0.986) were high. The SATS method presented an overall positive counting error (p<0.001) when compared to the SAFC, consistent with a fixed percentage overestimation of 11.2% (95% CI 8.3%, 14.2%) of the full count. The average time required to quantify axons by the SATS method was 10.9 min, only 27% of that required to conduct the SAFC.

CONCLUSIONS

Our data demonstrate that the SATS method provides a practical, rapid, and reliable means of estimating axon counts in the optic nerves of cats with glaucoma.

Longitudinal detection of optic nerve head changes by spectral domain optical coherence tomography in early experimental glaucoma

PURPOSE

We determined if the detection of spectral-domain optical coherence tomography (SDOCT) optic nerve head (ONH) change precedes the detection of confocal scanning laser tomography (CSLT) ONH surface, SDOCT retinal nerve fiber layer (RNFL), scanning laser perimetry (SLP), and multifocal electroretinography (mfERG) change in eight experimental glaucoma (EG) eyes.

METHODS

Both eyes from eight monkeys were tested at least three times at baseline, and then every 2 weeks following laser-induced chronic unilateral IOP elevation. Event and trend-based definitions of onset in the control and EG eyes for 11 SDOCT neural and connective tissue, CSLT surface, SDOCT RNFL, SLP, and mfERG parameters were explored. The frequency and timing of onset for each parameter were compared using a logrank test.

RESULTS

Maximum post-laser IOP was 18 to 42 mm Hg in the EG eyes and 12 to 20 mm Hg in the control eyes. For event- and trend-based analyses, onsets were achieved earliest and most frequently within the ONH neural and connective tissues using SDOCT, and at the ONH surface using CSLT. SDOCT ONH neural and connective tissue parameter change preceded or coincided with CSLT ONH surface change in most EG eyes. The SDOCT and SLP measures of RNFL thickness, and mfERG measures of visual function demonstrated similar onset rates, but occurred later than SDOCT ONH and CSLT surface change, and in fewer eyes.

CONCLUSIONS

SDOCT ONH change detection commonly precedes or coincides with CSLT ONH surface change detection, and consistently precedes RNFLT, SLP, and mfERG change detection in monkey experimental glaucoma.

Relationship between orbital optic nerve axon counts and retinal nerve fiber layer thickness measured by spectral domain optical coherence tomography

PURPOSE

We determined the relationship between total optic nerve axon counts and peripapillary retinal nerve fiber layer thickness (RNFLT) measured in vivo by spectral domain optical coherence tomography (SDOCT).

METHODS

A total of 22 rhesus macaques had three or more baseline measurements in both eyes of peripapillary RNFLT made by SDOCT. Laser photocoagulation then was applied to the trabecular meshwork of one eye to induce chronic unilateral IOP elevation. SDOCT measurements of RNFLT continued approximately every two weeks until the predefined study endpoint was reached in each animal. At endpoint, animals were sacrificed and the optic nerve was sampled approximately 2 mm behind the globe to obtain thin sections for histologic processing and automated axon counting across 100% of the optic nerve cross-sectional area.

RESULTS

At the final imaging session, the average loss of RNFLT was 20 ± 21%, ranging from essentially no loss to nearly 65% loss. Total optic nerve axon count in control eyes ranged from 812,478 to 1,280,474. The absolute number of optic nerve axons was related linearly to RNFLT (axon count = 12,336 × RNFLT((μm)) - 257,050, R(2) = 0.65, P < 0.0001), with a Pearson correlation coefficient of 0.81. There also was a strong linear relationship between relative optic nerve axon loss (glaucomatous-to-control eye) and relative RNFLT at the final imaging session, with a slope close to unity but a significantly negative intercept (relative axon loss((%)) = 1.05 × relative RNFLT loss((%)) - 14.4%, R(2) = 0.75, P < 0.0001). The negative intercept was robust to variations of fitted model because relative axon loss was -14% on average for all experimental glaucoma (EG) eyes within 6% (measurement noise) of zero relative loss.

CONCLUSIONS

There is a strong linear relationship between total optic nerve axon count and RNFLT measured in vivo by SDOCT. However, substantial loss of optic nerve axons (∼10%-15%) exists before any loss of RNFLT manifests and this discrepancy persists systematically throughout a wide range of damage.