Efficient determination of axon number in the optic nerve: A stereological approach

Induced Attenuation of Scleral TGF-β Signaling in Mutant Mice Increases Susceptibility to IOP-Induced Optic Nerve Damage

Purpose

Axonal optic nerve (ON) damage in glaucoma is characteristically associated with increased amounts of active transforming growth factor-beta 2 (TGF-β2) in the ON head. Here we investigated the functional role of scleral TGF-β signaling in glaucoma.

Methods

A deficiency of Tgfbr2, which encodes for TGF-β receptor type II (TGF-βRII), the essential receptor for canonical TGF-β signaling, was induced in fibroblasts (including those of the sclera) of mutant mice. To this end, 5-week-old mice were treated with tamoxifen eye drops. Experimental glaucoma was induced in 8-week-old mice using a magnetic microbead (MB) model. After 6 weeks of high intraocular pressure (IOP), the ON axons and their somata in the retina were labeled by paraphenylenediamine (PPD) and RNA-binding protein with multiple splicing (RBPMS) immunohistochemistry, respectively, and quantified.

Results

Tamoxifen treatment resulted in a significant decrease of TGF-βRII and its mRNA in the sclera. After 6 weeks of high IOP, reduced numbers of PPD-stained ON axons were seen in MB-injected eyes in comparison with not-injected contralateral eyes. Moreover, MB injection also led to a decrease of retinal ganglion cell (RGC) somata as seen in RBPMS-stained retinal wholemounts. Axon loss and RGC loss were significantly higher in mice with a fibroblast specific deficiency of TGF-βRII in comparison with control animals.

Conclusions

We conclude that the ablation of scleral TGF-β signaling increases the susceptibility to IOP-induced ON damage. Scleral TGF-β signaling in mutant mice appears to be beneficial for ON axon survival in experimentally induced glaucoma.

RGC-specific ATF4 and/or CHOP deletion rescues glaucomatous neurodegeneration and visual function

Endoplasmic reticulum (ER) stress has been linked with various acute and chronic neurodegenerative diseases. We previously found that optic nerve (ON) injury and diseases induce neuronal ER stress in retinal ganglion cells (RGCs). We further demonstrated that germline deletion of CHOP preserves the structure and function of both RGC somata and axons in mouse glaucoma models. Here we report that RGC-specific deletion of CHOP and/or its upstream regulator ATF4 synergistically promotes RGC and ON survival and preserves visual function in mouse ON crush and silicone oil-induced ocular hypertension (SOHU) glaucoma models. Consistently, topical application of the ATF4/CHOP chemical inhibitor ISRIB or RGC-specific CRISPR-mediated knockdown of the ATF4 downstream effector Gadd45a also delivers significant neuroprotection in the SOHU glaucoma model. These studies suggest that blocking the neuronal intrinsic ATF4/CHOP axis of ER stress is a promising neuroprotection strategy for neurodegeneration.

Differential effects of SARM1 inhibition in traumatic glaucoma and EAE optic neuropathies

Optic neuropathy is a group of optic nerve (ON) diseases with progressive degeneration of ON and retinal ganglion cells (RGCs). The lack of neuroprotective treatments is a central challenge for this leading cause of irreversible blindness. SARM1 (sterile α and TIR motif-containing protein 1) has intrinsic nicotinamide adenine dinucleotide (NAD⁺) hydrolase activity that causes axon degeneration by degrading axonal NAD⁺ significantly after activation by axon injury. SARM1 deletion is neuroprotective in many, but not all, neurodegenerative disease models. Here, we compare two therapy strategies for SARM1 inhibition, antisense oligonucleotide (ASO) and CRISPR, with germline SARM1 deletion in the neuroprotection of three optic neuropathy mouse models. This study reveals that, similar to germline SARM1 knockout in every cell, local retinal SARM1 ASO delivery and adeno-associated virus (AAV)-mediated RGC-specific CRISPR knockdown of SARM1 provide comparable neuroprotection to both RGC somata and axons in the silicone oil-induced ocular hypertension (SOHU) glaucoma model but only protect RGC axons, not somata, after traumatic ON injury. Surprisingly, neither of these two therapy strategies of SARM1 inhibition nor SARM1 germline knockout (KO) benefits RGC or ON survival in the experimental autoimmune encephalomyelitis (EAE)/optic neuritis model. Our studies therefore suggest that SARM1 inhibition by local ASO delivery or AAV-mediated CRISPR is a promising neuroprotective gene therapy strategy for traumatic and glaucomatous optic neuropathies but not for demyelinating optic neuritis.

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.

Predictors of Femtosecond Laser-Assisted Arcuate Keratotomy Efficacy for Astigmatism Correction in Cataract Surgery

PURPOSE

To evaluate the predictors related to the efficacy of epithelium- and Bowman membrane-penetrating femtosecond laser-assisted arcuate keratotomy (FSAK) in the treatment of preoperative astigmatism in patients with cataracts and to clarify the predictive role of the incision-to-limbus distance.

METHODS

This retrospective study included patients who underwent femtosecond laser-assisted cataract surgery combined with FSAK using the LenSX platform (Alcon Laboratories, Inc) at Beijing Aier-Intech Eye Hospital from March 2017 to December 2021. The Lenstar LS900 (Haag-Streit, Inc) was used to measure the horizontal corneal diameter, and ImageJ software (National Institutes of Health) was used to measure the vertical corneal diameter and incision-to-limbus distance. Finally, the measured values were converted. Changes in corneal astigmatism before and 3 months after surgery were analyzed using Alpins vector analysis. Correlation analysis and regression analysis were used to evaluate the factors associated with surgically induced astigmatism (SIA) in preoperative ocular biometric parameters.

RESULTS

The study included 94 eyes of 94 patients. The mean target induced astigmatism was 1.36 ± 0.44 diopters (D), SIA was 0.82 ± 0.43 D, and mean difference vector was 0.70 ± 0.40 D. Pearson correlation analysis and univariate regression analysis showed that preoperative corneal astigmatism, arcuate keratotomy arc length, incision-to-limbus distance, and astigmatism type were significant predictors of SIA. Multiple variable regression analysis included parameters such as age, arcuate keratotomy arc length, and incision-to-limbus distance and established a multiple regression model of SIA (all P < .01).

CONCLUSIONS

The incision-to-limbus distance was a significant independent predictor of SIA, and inclusion of this parameter may further improve the accuracy of the nomogram. [J Refract Surg. 2022;38(8):480-486.].

Neuronal NMNAT2 Overexpression Does Not Achieve Significant Neuroprotection in Experimental Autoimmune Encephalomyelitis/Optic Neuritis

Optic neuritis, inflammation, and demyelination of the optic nerve (ON), is one of the most common clinical manifestations of multiple sclerosis; affected patients suffer persistent visual symptoms due to ON degeneration and secondary retinal ganglion cell (RGC) death. The mouse experimental autoimmune encephalomyelitis (EAE) model replicates optic neuritis and significant RGC soma and axon loss. Nicotinamide mononucleotide adenylyltransferases (NMNATs) are NAD⁺-synthetic enzymes that have been shown to be essential for axon integrity, activation of which significantly delays axonal Wallerian degeneration. NMNAT2, which is enriched in axons, has been proposed as a promising therapeutic target for axon injury-induced neurodegeneration. We therefore investigated whether activation of NMNAT2 can be used as a gene therapy strategy for neuroprotection in EAE/optic neuritis. To avoid the confounding effects in inflammatory cells, which play important roles in EAE initiation and progression, we used an RGC-specific promoter to drive the expression of the long half-life NMNAT2 mutant in mouse RGCs in vivo. However, optical coherence tomography in vivo retina imaging did not reveal significant protection of the ganglion cell complex, and visual function assays, pattern electroretinography, and optokinetic response also showed no improvement in mice with NMNAT2 overexpression. Postmortem histological analysis of retina wholemounts and semithin sections of ON confirmed the in vivo results: NMNAT2 activation in RGCs does not provide significant neuroprotection of RGCs in EAE/optic neuritis. Our studies suggest that a different degenerative mechanism than Wallerian degeneration is involved in autoimmune inflammatory axonopathy and that NMNAT2 may not be a major contributor to this mechanism.

A method to prepare large resin sections for counting myelinated axons in rodent CNS and PNS structures

We present a method that allows preparing histological sections from large blocks of nervous tissue embedded in epoxy resin. Resin-embedding provides excellent resolution especially for the myelin-rich white matter and is often being used for visualizing the myelinated axons in peripheral nerves. However, because of the limited penetration of the reagents, only very small tissue specimens can be processed in this way. Here, we describe a method that enables to embed large specimens and their sectioning on a standard sliding microtome. To process the large specimens, modifications in several steps of the processing technique had to be made. In this paper we demonstrate, that with this technique 1-3 μm thick transversal sections can be prepared from the resin-embedded specimens as large as rat brain hemisphere. Such a large section allows simultaneously: 1.) overviewing and delineating the gross anatomical structures, and 2.) observing the subcellular details at the highest possible optical magnifications. Such a large section with excellent resolution allows application of unbiased stereological methods and reliable quantification of very small objects within the area of interest.

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.