Exploration of Structural Optic Nerve Changes in Mouse Models of Retinal and Neuronal Degeneration with Optical Coherence Tomography

This study demonstrates that polarization-sensitive optical coherence tomography can detect longitudinal structural changes in the optic nerve head of mouse models for Alzheimer's disease and neuronal degeneration, establishing it as a potential preclinical biomarker for these pathologies.

The Gist

Imagine your eye as a high-tech camera. The most important part of this camera isn't just the lens or the film; it's the cable that carries the picture from the camera back to your brain. In medical terms, this cable is the optic nerve, and the place where it plugs into the back of the eye is called the Optic Nerve Head (ONH).

This paper is like a detective story where scientists use a special "super-microscope" (called Optical Coherence Tomography, or OCT) to look closely at this cable plug in mice. They wanted to see if the plug changes shape when the mice get sick with diseases like Alzheimer's or age-related eye problems.

Here is the story of what they found, broken down into simple parts:

1. The Mission: Looking for Clues in the "Plug"

Scientists know that in humans with Alzheimer's, the brain gets clogged with sticky gunk (called amyloid plaques). They suspected that this gunk might also travel down the optic nerve cable, causing the "plug" (the ONH) to swell or shrink. But until now, nobody had really looked at this plug in mice models of these diseases.

They decided to use four different groups of mice:

• The "Alzheimer's" Mice: Three different types of mice designed to develop brain gunk (5xFAD, PS19, and APP/PS1).
• The "Aging Eye" Mice: Mice missing a specific shield against rust (oxidative stress), which makes their eyes age faster (SOD1 model).

2. The Tool: The 3D Super-Microscope

Instead of just taking a flat photo, the scientists used a machine that builds a 3D map of the inside of the eye. It's like using a drone to map a city block instead of just taking a picture of the street. They could measure the exact volume and shape of the optic nerve plug over time.

3. The Big Discovery: The "Swelling and Shrinking" Cycle

When they watched the "Alzheimer's" mice (specifically the 5xFAD type) grow up, they saw a very strange pattern in the optic nerve plug:

• The Teenage Swell: Between 3 and 5 months of age, the plug actually got bigger. Think of it like a balloon inflating.
• The Adult Crash: After 5 months, the plug started to deflate rapidly. By the time the mice were 9 months old, the plug had shrunk significantly.

Why did this happen?
The scientists believe the initial "swell" might be the nerve trying to fight the disease or the brain gunk starting to pile up. The later "crash" (shrinking) is likely the nerve fibers dying off, similar to how a tree branch withers when it loses its leaves.

The Gender Twist:
Just like in humans, the female mice showed these changes more dramatically than the males. The female "Alzheimer's" mice had much larger plugs than the healthy female mice, suggesting their eyes were reacting more strongly to the disease.

4. Comparing the Suspects

The scientists compared the different mouse models to see which ones showed these changes:

• The Amyloid Mice (5xFAD & APP/PS1): These mice, which have the "sticky gunk" (amyloid), showed clear changes in their optic nerve plugs.
• The Tau Mouse (PS19): This mouse had a different kind of brain gunk (tau protein). Surprisingly, its optic nerve plug looked normal. This suggests that the "sticky gunk" (amyloid) is the real troublemaker for the eye, not the other type.
• The Aging Eye Mouse (SOD1): These mice, which represent age-related eye degeneration, didn't show the same dramatic swelling and shrinking. Their changes were much slower and harder to spot.

5. The "Aging" Control Group

To make sure they weren't just seeing normal aging, they looked at healthy mice that lived to be very old (almost 2 years). Even healthy mice saw their optic nerve plugs shrink a little as they got very old, but the "Alzheimer's" mice shrank much faster and more drastically.

The Bottom Line: A New Early Warning System

The most exciting part of this paper is the idea of using the optic nerve plug as a biomarker.

Think of a biomarker like a "check engine" light in a car. Right now, doctors often have to wait until the engine is already broken (the patient has memory loss) to know something is wrong. This study suggests that by simply scanning the optic nerve plug with a camera, we might be able to see the "check engine" light turn on years before the brain starts to fail.

In summary:
The scientists found that in mice with Alzheimer's-like diseases, the optic nerve plug swells up early and then shrinks away as the disease progresses. This happens much faster than in normal aging. This discovery gives researchers a new, non-invasive way to track how well new drugs are working to stop Alzheimer's, all by looking at the back of the eye.

Technical Summary

1. Problem Statement

The optic nerve head (ONH) is a critical anatomical structure involved in the transmission of visual information and is affected by various ocular and neurodegenerative pathologies, including glaucoma, diabetic retinopathy, and Alzheimer's disease (AD). While the ONH is a standard diagnostic target in human ophthalmology, it remains underexplored in preclinical mouse models of disease. Specifically, there is a lack of quantitative data regarding ONH structural changes in mouse models of AD (amyloid and tau pathology) and retinal degeneration (oxidative stress models). Current research often focuses on retinal layer thickness or vascular parameters, potentially missing early or distinct pathological signatures located at the ONH.

2. Methodology

Animal Models
The study utilized four distinct transgenic mouse models to investigate different aspects of neurodegeneration and retinal degeneration:

• 5xFAD: A model of amyloid-beta pathology in AD (investigated longitudinally from 3 to 9 months).
• PS19: A model of tauopathy (investigated at 39 weeks).
• APP/PS1: Another model of amyloid pathology (investigated at 51–54 weeks).
• SOD1 Knockout (-/-): A model of oxidative stress and age-related macular degeneration (AMD), investigated longitudinally from 22 to 74 weeks.
• Controls: Non-transgenic (ntg) littermates and wild-type C57BL/6J mice were used for comparison, including a long-term aging cohort (up to 24 months).

Imaging System

• Technology: High-resolution Polarization-Sensitive Optical Coherence Tomography (PS-OCT).
• Specs: 840 nm central wavelength, ~3.8 µm axial resolution, 80–83 kHz A-scan rate.
• Protocol: Mice were anesthetized (isoflurane), pupils dilated, and imaged bilaterally. Volumes consisted of 2000 B-scans (512 A-lines each) covering a ~1x1 mm field of view centered on the ONH.

Image Processing and Segmentation Algorithms
The authors developed and evaluated two distinct quantitative approaches for ONH analysis based on intensity contrast:

1. Cross-Sectional Area Analysis:
   • The Retinal Pigment Epithelium (RPE) boundaries (anterior and posterior) were segmented using polarization contrast (depolarization by melanin).
   • The retina was flattened based on the RPE segmentation.
   • Reflectivity maps were generated at specific depths: Anterior RPE, Posterior RPE, and 20 µm and 40 µm below the posterior RPE.
   • Thresholding: A statistical threshold was applied (Mean intensity minus 2.5 × Standard Deviation) to segment low-intensity regions corresponding to the ONH.
2. 3D Volume Quantification:
   • A 52 µm axial range (30 µm above to 20 µm below the posterior RPE) was selected based on optimal signal-to-noise ratio.
   • Segmented areas within this range were stacked to calculate total ONH volume.
   • A secondary analysis was performed on a larger 142 µm range (presented in supplements).

Statistical Analysis

• Longitudinal data (5xFAD, SOD1) was analyzed using mixed-effects models to account for bilateral eye measurements.
• Cross-sectional comparisons (PS19, APP/PS1) used ANOVA with Tukey's post-hoc tests.
• Benjamini-Hochberg correction was applied for transgenic vs. non-transgenic comparisons.

3. Key Contributions

• Novel Biomarker Identification: Demonstrated that ONH structural metrics (area and volume) are sensitive to disease progression in AD mouse models, offering a potential new preclinical biomarker.
• Methodological Framework: Established a robust, intensity-based segmentation pipeline for ONH quantification in mice that does not require manual tracing, making it applicable to large datasets and various OCT systems.
• Differential Pathology Insights: Differentiated between the effects of amyloid pathology (5xFAD, APP/PS1) and tau pathology (PS19) on the ONH, suggesting amyloid accumulation has a more pronounced structural impact on the ONH than tau aggregation in these models.
• Longitudinal Aging Data: Provided detailed longitudinal data on ONH volume changes in both disease models and healthy aging controls, revealing distinct trajectories of ONH atrophy.

4. Key Results

5xFAD Model (Amyloid Pathology)

• Longitudinal Trend: ONH volume and area increased from 3 to 5 months of age (likely due to developmental growth) followed by a significant decrease from 5 to 9 months.
• Genotype Differences: At 9 months, transgenic (tg) 5xFAD mice exhibited significantly larger ONH volumes and areas compared to non-transgenic (ntg) littermates.
  • Volume: tg animals were ~27% larger than ntg (p=0.0008).
  • Sex Differences: The effect was more pronounced in females (32% larger volume in tg females vs. ntg females) than males.
• Comparison with other AD models:
  • APP/PS1: Showed significant differences between tg and ntg animals (similar to 5xFAD), with larger absolute ONH metrics than 5xFAD.
  • PS19 (Tau): No significant differences were found between tg and ntg animals, suggesting tau pathology alone may not drive ONH structural changes detectable by this method.

SOD1 Model (Retinal Degeneration/Oxidative Stress)

• No significant differences were observed between SOD1 knockout and control mice regarding ONH volume or area.
• A slight, non-significant shrinking trend was observed in knockout mice from 33 weeks onwards, but the sample size and effect size were insufficient for statistical significance.

Aging in Controls

• In wild-type controls, ONH volume increased slightly from 3 to 5 months, then decreased significantly by 9 months (~35% reduction from peak).
• By 22–24 months, volume had decreased by an additional ~10%, totaling a ~37% loss from the 3-month baseline.

5. Significance and Conclusion

• Preclinical Utility: The study validates the ONH as a viable target for monitoring neurodegenerative progression in mouse models. The ability to detect genotype-specific differences in amyloid models (5xFAD, APP/PS1) but not tau models (PS19) provides insight into the specific mechanisms of retinal involvement in AD.
• Biomarker Potential: The observed ONH atrophy in AD models aligns with human clinical findings (e.g., cup-to-disc ratio changes), suggesting these metrics could serve as surrogate markers for axonal health and disease severity in preclinical drug trials.
• Technical Advancement: The proposed segmentation method is simple, relies on standard intensity contrast (accessible on most OCT systems), and can be applied retrospectively to existing datasets.
• Limitations & Future Work: The authors note that the initial volume increase in young mice may be confounded by axial eye length growth. Future refinements should include eye length correction. Additionally, while the study suggests neural tissue loss is the primary driver of volume reduction, correlative histology is needed to definitively separate neural degeneration from vascular changes.

In summary, this paper establishes a quantitative framework for ONH analysis in mice, revealing that amyloid-driven neurodegeneration causes distinct, measurable structural changes in the optic nerve head that can be tracked longitudinally using PS-OCT.